CN113168903A - Product recommendation system and method - Google Patents

Abstract

A wearable device, comprising: a memory for storing product codes for consumables, topical application products and/or body worn products and data representing respective product recommendations or data representing product recommendations available for acquisition; a product code reader for reading a product code from a product; one or more inertial sensors for obtaining motion data of a wearer of the device; a visual indicator responsive to the read product code reading for providing a visual indication of a product recommendation using the data stored in the memory. The wearable device further comprises a processor configured to: processing the motion data to determine a period of time that the wearer is in a sedentary position or other sedentary oligodynamic state; analyzing the occurrence time and the duration of the time period; and adjusting the recommendation for at least a portion of the product code accordingly, wherein the product recommendation varies according to the determined time period.

Description

Product recommendation system and method

Technical Field

The present disclosure relates to wearable devices and methods of providing product recommendations. The present disclosure also relates to monitoring systems, wearable devices, and methods that alert a user to adverse lifestyle-induced effects of the user's environment and/or sedentary behavior. The invention relates particularly, but not exclusively, to providing recommendations based on product content and personal biometric information of the consumer.

Background

Semiconductor nanotechnology and optical technology have made a significant contribution to people's lifestyle, especially by promoting miniaturization of hardware. The use of these techniques for sequencing and genotyping has led to the development of so-called "lab-on-a-chip" systems. Wherein, according to the biological problem or target gene to be solved, the primer or probe is designed correspondingly, and the two substances are more commonly called as 'biological markers'. The biomarkers are oligonucleotides such as DNA molecules, and can target certain genes or variations. The biomarker may also be, for example, an antibody or antigen. By applying or selecting different types of biomarkers in such systems, consumers can test their biological samples, DNA, RNA, proteins, etc. (extracted locally or off-site by a third party, e.g., from saliva, blood, urine, tissue, feces, hair, etc.) for specific features that may be oracle for certain lifestyle issues or interests.

Such "personal" genetic or biological information may make medical decisions more effective, for example, by selecting a more likely effective treatment or drug dosage for a particular patient. Furthermore, by discerning differences between individuals at the molecular level, custom-tailored lifestyle and dietary recommendations based on the different needs of the individual or a particular population can be achieved. For example, personal care products such as cosmetics and nutraceuticals may be selected based on the availability of such products to individuals having a single nucleotide polymorphism in their DNA. To meet the ever-increasing consumer genetics market, a large number of private companies have been introduced and new genetic traits are characterized each day, thus continuously increasing the variety of biomarkers that have the potential to provide insight into the health and well-being of the general public, as well as genetic variations, phenotypes, etc.

Personal activity monitoring devices provide a means for users to conveniently record their physical activity. In particular, many so-called "fitness trackers" are capable of providing a user with the ability to estimate the distance traveled or run, the total energy expended, etc. Such data enables the user to make more informed decisions about lifestyle. In some cases, the fitness tracker may alert the user to a moderate or intense physical activity that has not been performed for the last period of time. While such fitness trackers may help users improve their health by encouraging them to perform more physical activity, other factors such as the diet they consume may also play an important role in determining the health and well-being of an individual.

WO2017055867 describes a wearable device that provides product recommendations based on biological information such as genetic data of a user. The wearable device includes a laser scanner or barcode reader for the wearer of the device to identify products that the wearer is interested in purchasing or consuming. The device then gives an indication of whether to recommend the product to the wearer based on the biometric information of the wearer. For example, the results of the DNA analysis may indicate that the user has a slower rate of caffeine metabolism than normal. In this case, the wearable device may recommend that the user avoid drinking coffee.

The effectiveness of product recommendations based on biological (genetic) and/or physiological information of a user in bringing benefits to the health of the user may vary from one user's behavior to another. Therefore, in order to achieve an improvement in the health condition of the user, it is necessary to improve the effectiveness of product recommendations.

Disclosure of Invention

The independent claims are directed to various aspects of the present invention. Further aspects and preferred features of the invention are set out in the dependent claims.

In this context, the term "consumable" primarily refers to products that are consumed orally, such as foods, beverages, supplements, pharmaceuticals and the like, but the term also covers products that are consumed transdermally. The term "topical application" refers to application to the outside of the body, such as skin or hair. Products that are applied (or can be applied) externally include, for example, cosmetics, creams, powders, or liquids.

Also disclosed herein is a computer-implemented system for providing recommendations to a user regarding consumables or topically applied (i.e., topically applicable) products, at least a portion of which is a body-worn portion. The system comprises:

a data storage device for storing a product code and data derived from personalized biological information obtained by analyzing a biological sample provided by a user;

a reader for reading or otherwise obtaining a product code from a product or product packaging;

a processor for obtaining a product recommendation for a product from the read or otherwise obtained product code and data stored within the data storage device, wherein the data stored within the data storage device includes at least data derived from personalized biometric information;

a user interface for providing an indication of a product recommendation to a user; and

one or more sensors for acquiring data representative of one or more physiological and/or biochemical functions of the user or representative of the user's environment, wherein the processor is configured to adjust one or more of the product recommendations based on the acquired data such that a modified indication is provided to the user via the user interface.

The body worn portion may have at least the reader and the one or more sensors disposed thereon, or the entire system may be the body worn portion. The body-worn portion may be a wrist-worn portion including a bracelet. The one or more sensors may include an accelerometer.

The processor may be configured to determine a value (e.g., number of steps) indicative of a user's activity based on data provided by the accelerometer, and the one or more product recommendations are adjusted based on the value of the activity. The adjustment may be applied to product recommendations that take into account the product calorie content. The one or more sensors may include one or more of a gyroscope, a heart rate monitor, a bodily fluid or chemical sensor (optionally including microneedles).

The personal biological information may be personal genetic information.

The processor may be operable to: storing, in the data storage device, a history of data representative of one or more physiological and/or biochemical functions of the user, for example a history covering a predetermined previous time period; and using the record to adjust one or more of the product recommendations such that the adjustment takes into account a past history of one or more physiological and/or biochemical functions of the user.

The data storage device may be used to store information relating to product content, including for example the carbohydrate content and/or sugar content of the product.

The product recommendation may have a first state of "recommended" and a second state of "not recommended", and the adjustment causes the product recommendation to change between the first and second states. The product recommendation may have a first state of "recommended," a second state of "not recommended," and a third state of "perhaps recommended," and the adjustment causes the product recommendation to change between the first and third states.

The user interface may be used to provide an indication of the product recommendation by illumination in different colors, red and green or red, green and amber.

The computer-implemented system may include another user interface for receiving an adjustment value from a user, wherein the processor is configured to adjust the scaling of the adjustment degree of the one or more product recommendations based on the adjustment value.

The reader may be a bar code scanner.

Also described herein is a computer-implemented body-worn system for providing recommendations to a user regarding consumables or topically applicable application products. The system comprises:

a data storage device for storing a product code and data derived from personalized biological information obtained by analyzing a biological sample provided by a user;

a reader for reading or otherwise obtaining a product code from a product or product packaging;

a processor for obtaining a product recommendation for a product from the read or otherwise obtained product code and data stored within the data storage device including at least data derived from personalized biometric information;

a user interface for providing an indication of a product recommendation to a user;

one or more sensors for acquiring data indicative of one or more physiological and/or biochemical functions of the user or indicative of the user's environment, wherein the processor is configured to adjust one or more of the product recommendations based on the acquired data such that a modified indication is provided to the user via the user interface.

The computer-implemented body worn system may include a bracelet.

Also disclosed herein is a computer-implemented method for providing recommendations to a user regarding consumables or topically applied (i.e., administrable) products, the method comprising:

storing in a data storage device a product code and data derived from personalized biological information obtained by analyzing a biological sample provided by a user;

reading or otherwise obtaining a product code from a product or product packaging;

obtaining product recommendations for a product from the read or otherwise obtained product code and data stored in the data storage device including at least data derived from personalized biometric information;

providing an indication of a product recommendation to a user via a user interface; and

data representative of one or more physiological and/or biochemical functions of the user or representative of the user's environment is acquired from one or more sensors, and one or more of the product recommendations are adjusted based on the acquired data, such that a modified indication is provided to the user via the user interface.

While one or more sensors are mentioned above for measuring one or more physiological functions of the user, the system or device may alternatively or additionally include one or more sensors for determining the location of the user or one or more environmental factors (such as pollution levels (e.g., NOx or particulate matter)) or Ultraviolet (UV) light levels to which the user is exposed.

Also described herein is a computer-implemented system for providing recommendations to a user regarding consumable or topically applicable application products, at least a portion of the system being a body-worn portion. The system comprises:

a product code reader;

one or more sensors for acquiring data representative of one or more physiological and/or biochemical functions of a user or representative of a user's environment; and

a processor for determining a product recommendation for the product identified by the product code reader based on the user's personal biometric information and the data acquired with the sensor.

The system can adjust the biology-based recommendations based on sensor output to gently persuade or encourage/discourage the use of certain products. The degree of the adjustment may be adjusted by the user, that is, the degree of influence of the sensor data on the biology-based recommendation is changed.

Also described herein is a computer-implemented method comprising: determining interception values of various nutrient components according to personal biological information of individuals; modulating or adjusting the cutoff value according to the current or recent physiological or biochemical function (e.g., activity) of the individual; and providing a product recommendation by applying the adjusted intercept value to a product such as a consumable or a topically applicable product.

Drawings

Fig. 1 is a perspective schematic view of a wearable device according to an embodiment of the present invention.

Fig. 2 is a system diagram of the wearable device of fig. 1.

Fig. 3 is a flow chart of data processing implemented by the wearable device of fig. 1.

Fig. 4A and 4B are flowcharts of data processing performed by the activity classifier of fig. 3 (fig. 4B is a continuation of fig. 4A).

Fig. 5 is a flow chart of data processing implemented by the lifestyle classifier of fig. 3.

FIG. 6 is a flow diagram of data processing implemented by the penalty state controller of FIG. 3.

FIG. 7 is a flow diagram of a method of providing product recommendations to a user.

Fig. 8 is a flow chart of data processing implemented by the wearable device of fig. 1.

FIG. 9 is a flow chart for adjusting product recommendations based on predicted user calorie sensitivity and user activity data;

FIG. 10 is a schematic diagram of a closed loop system for adjusting product recommendations.

FIG. 11 is a schematic illustration of two graphical user interface elements.

FIG. 12 is a graph of calculated sedentary time versus percentage of selected products for adjustment.

FIG. 13 is a table of graphical user interface elements updated according to a user's activities over a 24 hour period.

FIG. 14 is a schematic view of a graphical user interface including one of the graphical user interface elements of FIG. 11.

FIG. 15 is a schematic view of a graphical user interface including one of the graphical user interface elements of FIG. 11.

Detailed Description

By analyzing the genetic characteristics (genes) of the user, the risk or possibility of the user suffering from chronic diseases of long term, such as obesity, type 2 diabetes, cardiovascular diseases and the like, can be judged. Such genetic risks are fixed variables that cannot be adjusted. However, several adjustable factors, such as diet and physical activity, are available to reduce the risk of chronic disease to the user.

Embodiments described herein are directed to solving the above-mentioned problems by modifying the above-mentioned product recommendations by adjusting with measurements indicative of the physiological function of the user, for example measurements indicative of the user's last week caloric expenditure status or the user's heart rate data. By taking into account other factors (e.g., non-genetic factors) that may increase the risk of chronic disease, the user may be enabled to select products that are more likely to benefit their health.

For example, personalized food recommendations may be provided based on both the genetic profile of the individual and the physical activity level measured by sensors such as accelerometers. In addition, personalized product recommendations for other types of products such as cosmetics, pharmaceuticals, drugs, vitamins, etc. may also be obtained.

The gene testing service (provided by gene power company (DnaNudge) of london, england) provides personalized food recommendations based on individual genetic genes. Wherein several Single Nucleotide Polymorphisms (SNPs) of a subject individual are evaluated by DNA (or RNA) detection. Through scientific literature such as genome wide association research (GWAS), the SNP is associated with various chronic diseases such as obesity, type 2 diabetes and the like. The results of the gene detection are classified into five grades-extremely low risk, medium risk, high risk and extremely high risk. Subsequently, the gene detection result of the tested individual is associated with six nutrient substances of calorie, fat, saturated fat, carbohydrate, sugar and salt, and a set of nutrient interception values are obtained according to the associated result. Such nutrient cut-off values form the basis for personalized food recommendations. For example, if the salt content level of a certain product exceeds the corresponding nutrient cut-off value, the user is not recommended to consume the product.

Such "on the spot" personalized food recommendations may be provided to the user, for example, by a wearable device such as a bracelet device (referred to as "DnaBand"). The wearable device may also monitor the physical activity of the wearer and may determine one or more physical activity factors that reflect the amount of physical activity performed by the wearer while the device is worn. The level of physical activity of a person is taken into account in their baseline genetic recommendation. For example, the physical activity factor may be combined with a nutrient cutoff value for updating personalized food recommendations. Such recommendations are more targeted to the user, since they take into account that both diet and physical activity have an impact on the risk of chronic diseases.

One type of physical activity factor is the "calorie cutoff value," which is used to adjust the calorie nutrient cutoff value. For example, if it is determined that the user's physical exercise level is relatively low over a period of the past week or the like, a relatively low "calorie cutoff value" may be generated. If the calorie cutoff value is below the calorie nutrition cutoff value (determined from the genetic test results), a product recommendation can be generated based on the relatively lower calorie cutoff value therein. For example, a higher calorie nutrient cutoff may be set for users who are genetically non-obese. However, if the user has not recently performed too much exercise, the value may be reduced accordingly, recommending a higher calorie product (e.g., a bale of potato chips) as not suitable for the user.

Thus, by providing feedback of personal physical activity data with the wearable device, the personal calorie cutoff value may be adjusted. If the physical activity is insufficient, the calorie cutoff value will be reduced, and after the physical activity becomes sufficient again, the pre-reduction calorie cutoff value is again brought to baseline. The combination of physical activity, diet and genes forms a closed-loop feedback system, and more accurate personalized food recommendation can be provided. The user can control the degree of feedback to vary the amount of adjustment recommended for the product based on physical activity. For example, the user may not want the product recommendation to be affected by any physical activity, in which case the adjustment amount is set to zero. In addition, the user may want the above-described influence to be extremely large, in which case the above-described adjustment amount is set to a high value. The user can control the adjustment amount through some devices of the bracelet or through the interface of a computing device such as a smart phone.

Although lack of exercise has long been considered detrimental to health, the negative effects on sedentary behavior are often not appreciated. Ekeland et al, The Lancet 2016, 1303-. However, even with the recommended amount of daily activity, the user is not always able to do so, more often three times the recommended amount of daily activity. Setting the wearable device to adjust the product recommendations to the sedentary behavior of the device wearer may mitigate the effects of the sedentary behavior to some extent. In addition, many users may themselves act as an incentive to avoid sedentary behavior and engage in physical activity by clearly and visibly indicating that they are no longer recommending certain products because they are sitting too long.

The apparatus and methods presented herein stem from the recognition that: by using a wearable computing device to provide product recommendations that are tailored to the individual's sedentary behavior detected by the device, the individual's health condition may be improved.

Fig. 1 shows a wearable device 100 (or "bracelet") comprising a wristband 101, which in this example has a retractable portion 102 to enable a user to easily slide the bracelet 100 onto his wrist. In other examples, the wristband with the extendable portion may be replaced with a wrist band such as a watch wristband. Fig. 1 also shows three orthogonal axes X, Y, Z for describing the orientation of bracelet 100. The Y-axis is oriented along the axial direction of the bracelet, i.e. the direction in which the wearer's wrist penetrates the bracelet 100. The X-axis and the Z-axis are perpendicular to the Y-axis (and the X-axis and the Z-axis are perpendicular to each other), wherein the direction of the Z-axis is a direction of the bracelet 100 from top to bottom, i.e. the Z-axis is directed from the outside of the wrist to the inside thereof when the bracelet 100 is worn in such a way that the stretchable and contractible portion 102 is inside the wrist.

The bracelet 100 includes an optical sensor 103 such as a photovoltaic cell or a camera, and a light source 104 such as a laser. A window 105 is provided on the bracelet 100 so that the optical sensor 103 can be used to read the product code and the light source 104 can be used to illuminate the product code. In addition, an indicator (or indicators) such as a Light Emitting Diode (LED)106 is provided to provide product-related feedback to the wearer. The wrist band 101 is thicker on one side of its cross-section to accommodate various components of the wearable device 100 (see below).

Fig. 2 is a system diagram of the bracelet 100. Bracelet 100 is powered by a battery 201 that can be charged by a charging unit 202 and contains inertial sensors, accelerometers 203 (e.g., tri-axial accelerometers), and/or gyroscopes 204 for measuring movement of bracelet 100 in three-dimensional space. The axes of the accelerometer 203 and/or gyroscope 204 are aligned with the three orthogonal axes X, Y, Z shown in fig. 1.

Bracelet 100 may further include other sensors and electrodes 205, such as a heart monitor (e.g., an Electrocardiogram (ECG)) or thermometer for measuring a user's heart rate or temperature, and/or a GPS sensor (or other positioning system) for tracking a user's location. The sensors may for example comprise a microphone or an optical sensor for measuring heart rate.

Bracelet 100 is controlled by a processing unit 206 that accesses instructions and data stored within a memory 207. The wireless communication module 208 is used to enable the processing unit 206 to communicate with other computing devices such as other bracelets, smartphones, smartwatches, or personal computers. The wireless communication module 208 may, for example, be used to provide or update a product code database and/or a product recommendation database stored within the memory 207. Wireless communication module 208 may enable data exchange between bracelets 100.

The wearer may activate the light source 104 and cause light to be shone on the product to read the product code (or other information) of the product through the optical sensor 103. The wearer can manipulate or grasp the product with one or both hands to orient the product for proper reading. Alternatively, the product code may be read by the wearer by moving or aiming the bracelet 100 while the product is still on the supermarket shelf (for example). After reading the product code, the bracelet may provide feedback to the wearer through an indicator (LED)109, which may be an RGB LED that is adjustable to display different colors by mixing RGB components in different combinations and/or adjusting the brightness and/or blinking pattern of the LED. Of course, other types of indicators such as an array of LEDs or a screen, such as an LCD, LED or OLED screen, may also be used.

The sensors provided in the wearable device are, for example:

inertial sensors, such as accelerometers (e.g., three-axis accelerometers) and/or gyroscopes;

pedometer/step counter;

pulse rate sensors, such as photoplethysmography (PPG) sensors;

a respiration rate sensor;

heart rate sensors (also for measuring heart rate variability);

a blood pressure sensor;

microneedles for in situ blood testing, e.g. for blood glucose levels;

air mass or pollution sensors (e.g. mass spectrometry);

ultraviolet monitors (e.g., photodiodes).

Indicators 209, such as Light Emitting Diodes (LEDs) and/or vibrators 210, are used to provide visual or tactile feedback to the user wearing bracelet 200. In one example, the indicator 209 may provide product recommendations in accordance with a "traffic light system," where a "red" color indicates that the user is not recommended to use the associated product and a "green" color indicates that the user is recommended to use the associated product. In addition, an "amber" color may indicate that if the product recommendation is not adjusted based on the user's physical activity (e.g., sedentary or other physiological function), the product will be recommended as appropriate for the user. It should be understood that the various colors mentioned herein are not intended to be limiting and that other two-gear or three-gear (or multi-gear) indication recommendations may be used in addition thereto. For example, the bracelet 100 may display a representation of product nutritional information as well as nutritional cutoff values and/or modified cutoff values determined for the user.

Hereinafter, the respective control steps of the wearable device 100 providing product recommendations to the wearer will be described. In particular, the various steps of generating updated product recommendations will be described below, with sedentary behavior monitoring or physical activity status monitoring as examples. However, it will be appreciated that the techniques described below may be applied to other types of monitoring relating to one or more physiological and/or biochemical functions of a user or user environment.

Sedentary behavior

Fig. 3 shows how the motion data 301 collected by the inertial sensors 203, 204 is processed to update the product recommendations. In this example, motion data 301 includes acceleration components (x, y, z) of wearable device 100 along three orthogonal directions in three-dimensional space that are measured by tri-axial accelerometer 203 in a series of time steps (τ). Motion data 301 may further include data derived from the acceleration component, such as a position or velocity of wearable device 100. The motion data 301 may also include the orientation (or angular velocity or angular acceleration) of the device in three-dimensional space, as determined by the gyroscope 204, for example.

The motion data 301 is typically provided to the activity classifier 302 at a sampling rate of 40 Hz. As described in further detail below with reference to fig. 4A and 4B, activity classifier 302 determines an activity classification of the device wearer, such as "sedentary", "inactive", "active", or "unknown", by processing motion data 301. The activity classification may be updated upon receipt of each motion data 301 sample, or more commonly, upon receipt of a threshold number of motion data 301 samples. The latter case makes the active class update frequency lower than the above-mentioned sampling rate.

The activity classifier 302 provides activity classification data 303 to the lifestyle classifier 304, typically at a sampling rate of 1 Hz. As described in further detail below with reference to fig. 5, the lifestyle classifier 304 accumulates activity classification data 303 and determines a cumulative penalty list 305 for the data based on the device wearer's behavior. The cumulative time penalty is, for example, 24 hours.

The lifestyle classifier 304 typically provides the penalty list 305 to the penalty status controller 306 at a once-a-day frequency, but the interval may be shorter than a day (e.g., once an hour), or longer than a day (e.g., once a week). As described in further detail below with reference to fig. 6, penalty status classifier 306 determines a penalty status of wearable device 100 from penalty list 305. Further, the penalty status may also be determined based on other factors, such as age, gender, or product purchase history of the wearer (in some embodiments, product purchase history may be measured by wearable device 100), among other information. When the penalty state controller 306 determines that the device 100 should be placed in a penalty state, it will set a flag in the memory 207 to indicate that the device 100 should provide an adjusted product recommendation. If the penalty status is not determined, then the flag may not be set so that the device 100 provides product recommendations that are not affected by the wearer's sedentary behavior.

Fig. 4A shows a manner of processing of motion data 301 from the inertial sensor by a motion classifier 302 (referred to as a Motion Processing Unit (MPU) in the figure). Once wearable device 100 starts up, activity classifier 302 begins receiving sample 401 from the MPU. Each time a sample is received, the magnitude of the acceleration (e.g., euclidean norm) will be calculated from the x, y, z acceleration components measured by accelerometer 203. After receiving a threshold number of samples 403 (in this case, 40 samples), the variance 404 of the acceleration of the samples is calculated. It is then determined whether the device wearer is in a relatively stationary state by comparing 405 the variance to a threshold. If the threshold is not exceeded, another comparison is made to determine if the variance exceeds another threshold associated with the expected variance for slow walking. If the latter threshold is exceeded, the activity type is classified as "height" 407. Otherwise, the activity type is classified as "moderate".

Returning to the comparison 405 of the acceleration variance to the stationary behavior threshold, if the variance is less than the threshold, the sample data is filtered and processed 409 to determine if the device wearer is sedentary. The filtering process removes unwanted noise from within each of the x, y, z acceleration components, for example, with a band pass filter. The filtered acceleration component can be utilized in a variety of ways to determine whether the device wearer is sedentary. For example, the acceleration of the gravity contribution in the filtered acceleration component may be used to determine the forearm posture of the bracelet wearer (assuming the bracelet is worn in a conventional manner), and thus whether the wearer is engaged in a standing or sitting position for low activity. In the exemplary formula shown in element 409 of the figure, it is determined whether the acceleration of the device 100 along one axis (in this case, the X direction) is greater than the accelerations along the other two axes by comparing the filtered components with each other, which is generally related to the behavior of the wearer performing a low level activity in a standing or unsetting state (e.g., the wearer waving his arms back and forth while walking). The details of the above comparison operation in an implementation depend on the orientation of the accelerometer 203. The accuracy of sedentary detection can be improved by using a combination of gyroscope 204 and accelerometer 203. In addition, other sedentary detection methods may also be used, including machine learning methods that train classifiers based on motion data that have been labeled according to the type of activity the device wearer performs over time. From this analysis, the type of activity within the time period covered by the motion data sample may be determined as "sedentary" 410 or "low" 411.

After the activity type is determined, a corresponding one of the three counts is incremented according to the activity type. If the activity type is "high" or "moderate," the moderate to severe physical activity (MVPA) count 412 is increased, while if the activity type is "sedentary" or "low," the sedentary count 413 and the low count 414 can be increased, respectively. In addition to the activity type, a prediction index count is incremented 415 to count the number of predictions for the activity type. If the prediction index does not exceed the threshold value (in this case, 60 times), the activity classifier 302 waits for further reception of samples from the inertial sensor (MPU)401, and repeats the above-described processing procedure. The threshold is chosen to be a value that enables a reliable prediction of the wearer's behavior.

Fig. 4B is a continuation of fig. 4A and shows the processing performed by activity classifier 302 after finding a prediction index that exceeds the above threshold (i.e., after performing a sufficient number of classifications of activity types). Wherein a determination 417 is made whether the MVPA count is greater than a certain value (in this case, 30). If "yes," the lifestyle type associated with the movement data 301 sample is set to "active" 418. Otherwise, a determination 419 is made whether the low count is greater than a value (30 in this case). If "yes," the lifestyle type is set to "inactive" 420. Otherwise, a decision 421 is made whether the sedentary count is greater than a certain value (in this case, 30): if "yes," then set the lifestyle type to "sedentary"; if "no," the lifestyle type is set to "unknown". In other words, the final stages of the processes 417-423 are used to determine the primary behavior of the device wearer over the relevant time period, i.e., whether the user is active or sedentary for more than half of the time. After that, each count is reset, and the processing shown in fig. 4A and 4B is repeated.

Fig. 5 illustrates how lifestyle classifier 304 processes the lifestyle types derived from activity classifier 302. Lifestyle classifier 304 has an "active" (AC) count, an "inactive" (IC) count, or a "sedentary" (SC) count, and the initial value of these counts is set to zero 501. The initial state of classifier 304 is set to "non-sedentary". Upon receiving the lifestyle type (referred to herein as "1 st") 503 from the activity classifier 302, the lifestyle classifier 304 determines its type and increments the corresponding count while setting the other count values to zero 505, 506, 507. For example, when the count of "active" is incremented, then the count of "inactive" and "sedentary" are set to zero 505. Upon incrementing the "active" count value or the "inactive" count value, the state of classifier 304 remains "non-sedentary" and processes 502-504 are repeated for the next lifestyle type determined from motion data 301. If the count value of "sedentary" is incremented 507, it is determined whether the "sedentary" count value has reached a particular value 508 (in this case, 2). That is, steps 504-508 are used to determine whether the device wearer has continued sedentary (i.e., sedentary uninterrupted by any MVPA or low-level non-sedentary activity) for a length of time (2 cycles). If this criterion 508 is not met, the state of classifier 302 remains "non-sedentary" and the process described above is repeated again. However, if the evaluation criteria 508 are met, then the "active" and "inactive" count values are set to zero 509, and the state of the classifier 304 is set to "sedentary".

Once lifestyle classifier 304 enters "sedentary" state 510, it will continue to receive lifestyle types from activity classifier 302. In the "sedentary" state, lifestyle classifier 304 processes lifestyle types 511, 513 in a similar manner as in the "non-sedentary" state. However, each time the "active" count value 513 or the "inactive" count value 514 is incremented (and the other count values are set to zero), it is determined whether the "active" count value or the "inactive" count value exceeds a certain value (2 in this case). That is, the classifier 304 determines whether the device wearer has continued MVPA in the first two cycles or has continued low non-sedentary activity in the first two cycles. If so, the "active" and "inactive" count values are reset to zero 516 and the state of the classifier 304 is restored to "non-sedentary" 502. Otherwise, the "sedentary" count is incremented 517.

After incrementing the sedentary count, in either step 517 or 518, a determination is made as to whether the "sedentary" count exceeds the threshold 520 (in this case, 30). If "yes," then the penalty is added to the penalty list 520 and the "sedentary" count is set to zero 521. Subsequently, the process is restarted 510, while the state of the classifier 304 is still "sedentary".

Figure 6 illustrates the manner in which the penalty state controller 306 controls the behavior of the wearable device 100 using the penalty list. When the device 100 is started or reset 601, the controller 306 determines if a certain length of time (in this case, 24 hours) has elapsed since the last operation of the wearable device. If "yes," the state of the controller 306 is set to "green" and the penalty score is set to zero 604 at the same time. In this "green" state, no adjustment of the product recommendation is made when the device 100 is used to read the product code, but rather a "normal" product recommendation (e.g., based on the user's genotype) is presented to the user.

Controller 306 waits 605 to receive a new penalty from lifestyle classifier 304. When a penalty is received, the penalty score is incremented 607 and a determination 608 is made as to whether the "sedentary" criteria rule described above is satisfied. These rules may vary depending on user information, such as whether the user is a child or adult. For example, the criteria for an adult may be: if the penalty score is 12 or more over a 16 hour period, the user is classified as "sedentary". For children, the penalty score required by the corresponding rule may be only 6 or more. If the sedentary criterion is not met 609, then it may be checked whether a new day is entered 609 (or whether some other period of time has elapsed). If "no," the controller 306 continues to wait 605 for further receipt of penalty point points from the lifestyle classifier 304. If a new day is entered, the controller's state is still "green" as described above, and the penalty score is set to zero 604.

When the sedentary evaluation rule(s) is not satisfied, the state of the controller 306 will change to "amber" 611. In this state, the product recommendations provided by the device 100 need to be adjusted. The controller 306 remains in this state until the next day 612 is entered. After the next day of entry, the penalty score is again set to zero 604 and the state returns to "green" (not shown).

In one example, product recommendations can be divided into three different types, "not recommended," perhaps recommended, "and" recommended. Wherein the user may be indicated the type using the color changing LED 106 of the device 100. For example, traffic light systems of three colors red ("not recommended"), amber ("perhaps recommended"), green ("recommended") may be used. When the controller 306 is in the "green" state, then a "normal" product recommendation is provided to the user. However, if the accumulated penalty points are excessive and the controller 306 enters an "amber" state, some of the product recommendations that would otherwise be indicated to the user in green will be indicated to the user in amber. For example, for a user, a food product having a relatively high calorie content may generally be recommended as a food product suitable for consumption (i.e., indicated in green), but if the user sits for an extended period of time without movement, the product recommendation may be adjusted (modulated) to be "perhaps recommended" (i.e., indicated in amber).

The controller 306 may also be used to set a portion of the products that should have been "perhaps recommended" (i.e., indicated in amber) as "not recommended" (i.e., indicated in red) based on the sedentary behavior described above of the device wearer. However, in general, "not recommended" products are not affected by sedentary behavior (i.e., always indicated in red).

The use of the three product recommendations "status" described above is for exemplary purposes only. Of course, only two product recommendation states ("recommended" and "not recommended") or more than three product recommendation states may also be used, such as a recommendation score of 100 full score, or two score values, where a negative value indicates a non-recommended product and a positive value indicates a recommended product. Where recommendation scores are used, the above adjustments may be made by reducing the recommendation score according to the number of penalty points accumulated by the wearer.

The portion of the products that make the recommended adjustments for sedentary behavior may be selected in a variety of ways. Typically, no adjustments are made to all types of food products, as this may make "healthy" products such as vegetables "perhaps recommended" due to the sedentary behavior of the user. The portion of products may be selected based on nutritional data associated with each product. In one example, for products classified as "recommended," the products are ranked according to their calorie content, and products with the top 50%, 30%, or 10% of the calorie content among the products are selected as products for adjustment. In some embodiments, the user may adjust the cut-out percentage of the products selected to increase or decrease the number of products for adjustment.

Generally, product recommendations can be obtained by the following procedure. First, several health characteristics that are likely to be associated with poor health conditions, including, for example, type 2 diabetes, hypertension, high Body Mass Index (BMI), and high cholesterol, are identified. Wherein it can be determined whether the user has any such health characteristics by genetic testing, e.g. based on the detection of Single Nucleotide Polymorphisms (SNPs) in a DNA sample of the user. In addition, other forms of diagnostic testing may be employed, such as breath testing for determining the microbiome composition of the user.

Once the "target" health characteristics are determined, the effect of the diet on each health characteristic is initially taken into account. This may be achieved by analyzing each of several different categories relating to the nutritional content of the consumable, including, for example, any of the following: sugar content, calorie content, carbohydrate content, saturated fat content, total fat content, and salt content. For example, it is known that consumption of high-sugar products increases the risk of type 2 diabetes, while a salt-rich diet is associated with hypertension. The relationship between each class and each healthy feature may be quantified by assigning a feature correlation coefficient to each class. Taking hypertension as an example, a larger factor may be assigned to the salt content and fat content, and a smaller factor (or even a zero factor) may be assigned to the calorie content and carbohydrate content.

For each health characteristic determined, a characteristic correlation coefficient may be used to calculate a product score that represents the expected impact based on the nutritional content of the product (i.e., how many grams of salt, saturated fat, etc. the product contains). For example, in calculating a score that represents the adverse effect of a product on hypertension, the salt content coefficient may be multiplied by the number of grams of salt in the product. In mathematical terms, the score for each feature can be determined by taking the scalar product of its feature correlation coefficient vector and the nutritional information vector for each product. Of course, more sophisticated score calculations may also be used, for example, the impact of different nutritional categories on a healthy characteristic may be modeled using a determined dose response curve for that characteristic. The score may also take into account other factors such as product type (e.g., candy, cookies, breakfast cereal, etc.) and typical serving size of the product. Such other factors may be used to adjust the score for differences in consumption patterns of different products. For example, if a product is determined to be generally consumed as a "snack," its score may be reduced, thereby making recommendations for that product more positive than would otherwise be given.

After the score for each product is calculated, the products are ranked in descending order of their score for a particular health feature. A portion of the products are then selected by selecting products that rank above a threshold rank. For example, the top 50%, 30%, or 10% ranked products may be selected based on their health feature scores. Such products may be assigned "negative" recommendations, such as "not recommended". As such, when the bracelet 100 reads the barcode of the product, the LED 104 turns red (for example). The remaining products are assigned other categories such as "recommended" (the user will be prompted in green). Thus, each product assigns different recommendations depending on its ranking for a particular feature. The threshold ranking value (or "cutoff value") used to assign recommendations varies from user to user and is determined by how susceptible the user is to the particular feature (e.g., as determined by genetic testing).

The potential impact of sedentary (or other form of sedentary oligodynamic behavior) is considered by selecting another portion of the product for which product recommendations are adjusted based on the penalty status of the controller 306. The portion of products may be determined by selecting products ranked below the threshold ranking and above a second threshold ranking. For example, for a particular feature, if 50% of the products are "recommended" products, the top 20% of those products may be selected as part of the adjusted product. The second threshold ranking is selected according to the severity of the impact of sedentary and/or other types of sedentary oligodynamic behavior on health characteristics. The portion of products may be assigned to a particular database table or otherwise "tagged" within the memory of the bracelet 100 to facilitate determining whether product recommendations for a particular product need to be adjusted. In one example, if the controller 306 is in an "amber" state, each product within the portion of products is prompted as a "PermitRemmable" (e.g., the indicator changes to amber). However, if the controller 306 is in the "green" state, then each product is prompted as "recommended" (e.g., the indicator changes to green).

The total product recommendation for each product is determined by combining the product recommendations for each health feature. This may be accomplished in a number of ways, such as determining a recommendation for the product as "not recommended" if there is a recommendation for any feature as "not recommended" or if there is a recommendation for more than one feature as "not recommended". Further, for each product, a flag indicating whether the product recommendation should be adjusted may be set or not set according to the same or similar rule as above.

In general, the above threshold rankings and the rules for determining total product recommendations are balanced in two ways: allowing the user to select the product that the user wishes to consume; users are discouraged from consuming at least a portion of the product that is most likely to have an adverse effect on their health. The advantage of this balance is that the user can be "gently persuaded" to make a choice that is more favorable for him in the long term. Adjusting product recommendations based on a user's sedentary and/or other sedentary behavior may provide the additional effect of "gentleness persuading" -for example, a user may find that a certain originally "recommended" product becomes a "perhaps recommended" ("self-determined") product after a long period of sedentary time within a certain day. The adjustment without adjusting the product to "not recommended" may result in the following particularly significant advantages: in certain situations where the user is not in control of himself, for example, where the user has to sit for a long distance, no overly severe penalty is imposed on the user.

FIG. 7 is an overview of the steps involved in a method of providing product recommendations to a user. The method comprises the following steps:

step 701: one or more health characteristics of the user, such as type 2 diabetes, hypertension, high Body Mass Index (BMI), and high cholesterol, are determined by genetic testing of a biological sample provided by the user.

Step 702: for each of a plurality of consumables, a score is calculated that represents the extent of the effect of the product on each of the health characteristics, each score based at least in part on nutritional information for the product. For each product, a score may be calculated, for example, by the process described above: determining the nutrient content of each of its several categories (e.g., via information provided by the manufacturer); multiplying the nutrient content of each category by a feature correlation coefficient; and obtaining a total score for each feature by adding the obtained values.

Step 703: based on the score, a product recommendation is assigned to each product. As mentioned above, this can be achieved by: ranking the products according to the corresponding scores of the products for each health characteristic; selecting a part of products from the sorted products aiming at each health characteristic; assigning a recommendation to each portion of the product; and obtaining a total product recommendation for each product by combining the recommendations for each health feature. The product recommendation may be selected from a set of predetermined product recommendations.

Step 704: the user's behavior is monitored by determining the period of time that the user is in a sedentary position or other sedentary oligodynamic state.

Step 705: product recommendations for at least a portion of the products are adjusted based on the user's behavior. The product recommendations may be adjusted, for example, by assigning different product recommendations from a set of predetermined product recommendations. The portion of products may be determined based on the product score for each health characteristic. For example, a certain percentage of the product that is "recommended" (i.e., "green") for a particular characteristic may be reassigned as "perhaps recommended" (i.e., "amber").

Step 706: the adjusted product recommendations are provided to the user via a visual indicator, such as the light source 104 of the bracelet 100.

Physical activity and environmental monitoring

Fig. 8 is a block diagram of how the inertial sensor 801 output signal in a wearable device is processed according to one embodiment. In the present case, the sensor provides signal data associated with motion about three orthogonal axes (x, y, z). The signal (along with signals from any other sensors within the wearable device) is then sampled. The output signal of the inertial sensor is down-sampled at least at 20Hz (i.e. at the nyquist frequency of 10 Hz) to obtain all signal content associated with moderate and high intensity physical activities/movements such as walking and running.

The sampled signal (i.e., the original signal) is then pre-processed 802 by filtering. For inertial signals, a band-pass (BP) filtering pipeline with a bandwidth of 0.25-8 Hz is adopted to eliminate the unwanted noise components and ensure that all signal components related to medium-intensity and high-intensity activities are reserved. Among them, a second-order Butterworth (Butterworth) filter is employed because such a filter can achieve a smooth transition between a pass band and a stop band and a uniform unity gain within the pass band. In order to ensure the stability of the filters, each filter adopts a zero-pole analysis design. In addition, the nonlinear influence of phase response is eliminated by a zero-phase filtering technology for sequentially carrying out forward and reverse filtering on signals.

The signal is analyzed in segments, where the length (duration) of each segment is configurable, but typically ranges from 1s to 60 s. For each segment or "window" of the preprocessed accelerometer data, its average size or "energy" is calculated 803. Such an average may be referred to as an "active accelerometer count value" (AAC). For example, when using a three-axis accelerometer, the measured acceleration components along each axis (which are typically corrected) can be summed and then averaged over different (discrete) time points within the data acquisition window, for example, using a numerical orthogonality Rule such as Simpson's Rule. Alternatively, the total acceleration may be calculated using the vector norm (i.e., the two-norm) of each component, and then averaged.

Subsequently, it is determined by a Physical Activity (PA) classifier 805 whether the user is in an active state of physical exercise within each time window. This can be achieved, for example, by applying a simple thresholding rule to the windowed AAC data, i.e. if the AAC value of a certain window exceeds a specified value, it is determined that the user is in an active state of physical exercise during this period of time. Wherein a suitable threshold value may be determined, for example, by measuring AAC values obtained by the user while performing different types of physical exercises at different intensity levels. In addition, more sophisticated classifiers may also be utilized to determine the intensity or type of physical activity that has been performed, for example to distinguish between moderate or extreme activities, or to distinguish between running and cycling activities.

The above classification of whether a user is in an active state of physical exercise is then used to classify the user into one of two categories, "inactive" and "active". The classification of a user as "inactive" or "active" is determined based on whether their activity classification values meet the requirements of evidence-based guidelines on physical activity, such as national institute of health and clinical optimization (NICE, a government organization in the uk). The guidelines generally give the desired amount of physical activity for different age groups.

For example, the NICE Physical Activity Guideline (PAG) recommendation for the age of 19-64 years:

adults should target daily exercise. The total duration of medium intensity activity of 10 minutes or more in duration is at least 150 minutes (2.5 hours) a week-one such activity being performed for 30 minutes per day for at least 5 days a week;

an alternative way to achieve the same beneficial effect is to perform 75 minutes of high intensity activity within a week, or to perform a combination of medium and high intensity activity;

an adult should perform physical activity that increases muscle strength for at least two days a week;

all adults should reduce the time to sedentary (sedentary) as much as possible.

In one implementation of the NICE guidelines, an individual is "inactive" if the person has a moderate physical activity for less than 150 minutes per week, or a high physical activity or a combination of moderate and high physical activity for less than 75 minutes. Further, if an individual meets the above requirements, the individual may be deemed "active" according to the guidelines.

Since the physical activity guide described above is based on weekly guidelines, the present closed loop feedback system is a dynamic system and changes personalized food recommendations depending on whether the user changes from "inactive" to "active" or from "active" to "inactive". Wherein changes in the average level of user activity may be taken into account by calculating a rolling average of the activity data, for example by classifying whether the user was "active" or "inactive" in the last week. Of course, other average times, such as 1 day or about 1 month, may also be used.

The resulting category (i.e., "active" or "inactive" in this example) is then passed to decision-making program 805 for it to use the category and other information (DNA and/or nutritional information) to determine a recommendation to update the product stored in the bracelet memory (e.g., from green to amber). Alternatively, the value of the "modifier" may be stored in the device (or stored remotely) and used immediately after the user scans the product to update the recommendation.

As described above, the physical activity category may be used to adjust a calorie cutoff value that is used to determine whether to recommend use of the product based on the calorie content of the product. The adjustment depends on whether the user is classified as "active" or "inactive".

Active-if an individual meets the requirements of a PAG, personalized food recommendations are based only on genetic genes, without modification to the recommendations.

Inactivity-if an individual does not meet the requirements of the PAG, only the calorie nutrient cutoff value needs to be adjusted. Wherein by increasing the individual's calorie sensitivity, the calorie cutoff is decreased, thereby correspondingly decreasing the allowable calories ingested.

In weight management, energy balance is one of the key factors. Energy may be measured in calories or kilojoules and is derived from the total amount of protein, fat and carbohydrates in the food. The key to long-term weight management is to ensure the correct balance between the number of calories consumed (input) and the number of calories utilized (output) by an individual.

Depending on the degree of energy balance, there may be three weight management situations: (1) if calorie intake is greater than total energy expenditure, weight will increase; (2) if calorie intake is equal to total energy expenditure, body weight will remain unchanged; (2) if calorie intake is less than total energy expenditure, weight will be lost. Thus, in order to prevent a state of weight gain (due to net calorie intake), it is desirable to regulate the calorie intake of an individual, either down-regulated from baseline, or up-regulated to baseline.

This regulation of calorie cutoff generally shifts several products from "green" recommendations to "amber" recommendations. In this case, the amber color indicates that a food product is not recommended due to lack of physical activity, but once physical activity is sufficient, the recommendation for the product turns "green". However, it is important that the amount of health foods such as vegetables recommended for people is not reduced. Therefore, the adjustment of calorie cut-off value is only applicable to certain kinds of food, such as potato chips, chocolate, candy, etc., which are classified as food "at the user's discretion".

In the following, various scenarios will be described in which product recommendations are adjusted based on different combinations of genetically determined user sensitivities or trends and measurement data.

FIG. 9 illustrates a process for adjusting product recommendations based on a user's predicted calorie sensitivity and a user's activity data. The process comprises the following steps:

a: the product is as follows: calorie content

B: and (3) DNA detection results: calorie sensitivity test results (all users)

C: and (3) real-time measurement: moving: number of steps/distance in a given time, and the result of comparison with a baseline

D: recommended adjustment: high calorie products are not recommended

E: reminding content: encourage more users to do exercise

The following examples are similar to the example described with reference to fig. 9, except that the steps in each example are replaced with the steps a to E, respectively.

In one example, the product recommendations are adjusted based on the predicted caffeine metabolic rate and the time of day of the user. The effect of caffeine is more durable for people with "slower" rates of caffeine metabolism. By measuring the time of day and/or the user's heart rate continuously in real time, adjustments can be made to whether a particular caffeine-containing product, such as a coffee or energy drink, is recommended.

A: the product is as follows: coffee, tea, sports drinks, protein milkshakes, carbonated beverages (cola)

B: and (3) DNA detection results: rs 762551-slow metabolism of caffeine SNP

C: and (3) real-time measurement: time of day

D: recommended adjustment: not recommending products

E: reminding content: reminding the user of the DNA detection result

In another example, the product recommendation is adjusted based on the predicted hypertension predisposition of the user and heart rate data of the user. For users with higher resting heart rates, they may be encouraged to eat, for example, fat-rich fish and nuts by adjusting the initial (i.e., "healthy state") fat recommendations in the category. Similarly, the nutrient cut-off of salt may be reduced and/or the administration of omega-3, omega-6, omega-24 supplements may be recommended. In addition, each interception value can be further adjusted according to the heart rate over time.

A: the product is as follows: total fat-but with a focus on product category, product salt and saturated fat content

B: and (3) DNA detection results: hypertension SNP

C: and (3) real-time measurement: heart rate

D: recommended adjustment: reduction of saturated fat content depending on the category-encouraging use of healthy fat-containing foods

E: reminding content: alerting the user to avoid the use of saturated fats/salts for heart health and encouraging the user to scan for products beneficial to the heart

In another example, the product recommendation is adjusted based on a measure of the amount of perspiration produced by the user. Wherein the recommendation for vitamins is adjusted based on the level of sweating.

A: the product is as follows: isotonic solution, protein milk shake, and other sports supplements

B: and (3) DNA detection results: all users

C: and (3) real-time measurement: amount of perspiration produced

D: recommended adjustment: encouraging the consumption of isotonic solution/recommending protein supplements

E: reminding content: encourages taking tonics when necessary

In yet another example, product recommendations are adjusted based on predicted user sun sensitivity and user UV exposure measurements. Wherein the exposure level can be determined by tracking the position of the user and knowing the ultraviolet exposure level of the user using the UV reference map. This method can also be used to determine the contamination exposure level of the user. Wherein additionally or alternatively an ultraviolet sensor may be used, e.g. a photodiode may be integrated within the wearable device. This information can be used to adjust SPF recommendations to recommend high protection sunscreens, for example, between high protection sunscreens and sunscreens with lower SPF values.

A: the product is as follows: with or without SPF (Sun protection factor)

B: and (3) DNA detection results: genes associated with sun sensitivity, e.g. NTM AA, TYR GG, ASIP TC, LOC10537 CC

C: and (3) real-time measurement: LED, ultraviolet/visible spectrum

D: recommended adjustment: varying SPF cut-off values

E: reminding content: if the user's ultraviolet exposure level is high, a reminder is given to the user

In another example, product recommendations are adjusted based on predicted vitamin E production by the user and measurements of the user's UV exposure. Ultraviolet light (and sun) can reduce vitamin E levels in the skin. Vitamin E absorbs energy from Ultraviolet (UV) radiation. The user's product recommendations can be adjusted to favor the vitamin E-promoting ingredients through uv maps (location based) or built-in uv measurement devices.

A: the product is as follows: with or without vitamin E

B: and (3) DNA detection results: SNP of vitamin E

C: and (3) real-time measurement: LED, ultraviolet/visible spectrum

D: recommended adjustment: altering vitamin E cut-off

E: reminding content: if the user's ultraviolet exposure level is high, a reminder is given to the user

In yet another example, the product recommendation is adjusted based on the predicted collagen degradation likelihood of the user and the skin hydration level measurement and/or skin oil level measurement (via a skin oil tester) and/or skin pH measurement of the user. Wherein the use of the product may or may not be recommended depending on the oil content and/or the pH balance of the product.

A: the product is as follows: all skin care products

B: and (3) DNA detection results: genes for undesirable collagen degradation (loss of water aggravates wrinkle formation)

C: and (3) real-time measurement: e.g. skin hydration levels as measured by a skin stratum corneum moisture content tester

D: recommended adjustment: encouraging the use of products containing moisturizers, blockers and emollients

E: reminding content: encouraging the user to achieve optimal skin health by maintaining hydration

In yet another example, product recommendations are adjusted based on a predicted likelihood of a user being negatively affected by contamination and a contamination exposure measurement of the user. The reason for this is that contamination can cause damage to the skin. The NQO1 gene has an effect on an individual's ability to tolerate environmental toxins. At present, the negative impact of PM2.5 (a fine particulate matter, an airborne mixture of fine solid particles and liquid droplets) is increasingly recognized, especially for urban inhabitants. The users of cosmetics pay more attention to pollution, and "anti-pollution" becomes a new field of the cosmetic industry. These types of products may be advantageously recommended for users who expect to be likely to have some tolerance to environmental toxins, but who themselves have been exposed to a very high degree of contamination.

A: the product is as follows: skin care products with or without anti-fouling ingredients

B: and (3) DNA detection results: NQO1 gene (Normal/poor contamination resistance)

C: and (3) real-time measurement: location detection and query of location-based pollution maps, or Particulate Matter (PM)_2.5) Detection of iso-particulate matter

D: recommended adjustment: recommendations to help resist contamination

E: reminding content: remind the user to pay attention to the DNA detection result/inform the user that the user is in a pollution exposure state

FIG. 10 is a schematic diagram of a closed loop method for providing product recommendations. Personalized genetic data (or other biologically derived data) 1001 is stored in database 1003. As described above, this data is used to generate cut-off or threshold values for different nutritional components such as carbohydrates, fat, salt, etc. These values are adjusted up or down by the regulator 1005 according to the physiological and/or biochemical (or environmental) function determined by the unit 1011 receiving sensor data from the wearable device 1007. Wearable device 1007 provides product recommendations 1009 using the adjusted intercept values and product data. Of course, all of the components shown in the figures may be disposed within the wearable device 1007.

Alerting a user to the effects of his or her behavior and/or environment

By providing a person with more effective feedback of the consequences of sedentary behavior, the person may be encouraged or "persuaded" to improve lifestyle. In particular, by visually indicating to the user that several product recommendations have been adjusted for their sedentary behavior, it helps to alert them to the fact that the effects of sedentary behavior should instead be offset by changing diet (for example). In this manner, the user may be motivated to reduce the number of products for which the respective product recommendation is adjusted by avoiding prolonged sedentary and physical activity.

Once the length of time that the user is sedentary in a seated position is determined, for example, as described above, it can be determined how the product recommendation should be adjusted by analyzing the length of time/sensor data. For example, if the user has been sedentary for 30 minutes (for example), the user may be given a "sedentary point". Generally, the sedentary time will be accumulated to determine the sedentary point number using the total sedentary time of the user. Alternatively, in some embodiments, a sedentary point may be assigned to the user only when the user is sedentary for more than a preset length of time and without a break time of more than 2 minutes (or some other shorter length of time). Subsequently, a product that requires adjustment of the product recommendation can be determined based on the sedentary point number. For example, products may be ranked according to a score that indicates an expected negative impact of the product on one or more particular health characteristics. For example, the products may be ranked according to their salt content in grams (or saturated fat, etc.) and their expected effect on hypertension (for example). In this case, each sedentary point may represent a product recommendation that requires an additional 10% of the ranked products to be adjusted, or "downgraded".

Another metric of sedentary behavior is called "calculated sedentary time" (CST), which increases with increasing sedentary time of the user and decreases after the user has performed physical activity. One form of this metric parameter can be expressed mathematically as:

CST＝(T_BLOCKtotal number-Total step/R_STEP)/max(T_BLOCK)

In the formula, T_BLOCKRepresents a sedentary point count, i.e., the amount of time that the user spends less exercise (e.g., sedentary) for a length of time that exceeds some threshold. In this example, the amount of physical activity performed by the user is divided by the number of steps of the user divided by a constant R_STEPTo quantify R_STEPIndicating the cancellation of a sedentary integral point (T)_BLOCK) The number of steps required (e.g. 1000 steps). The CST may be calculated once per hour. In a specific example, max (T)_BLOCK) This factor is used to ensure that the maximum value of CST per hour is 1. For example, max (T) when a half-hour interval is taken_BLOCK) Is 2. In some embodiments, T_BLOCKThe value may be set by the user.

The processing of the sensor data described above may be performed by the wearable device 100. Alternatively or additionally, some or all of the sensor data generated by the sensors of wearable device 100 may be sent to a personal computing device (e.g., a smartphone) for processing. For example, the personal computing device may determine the time the user is sedentary based on sensor data provided by wearable device 100.

FIG. 11 shows an example of a portion of a Graphical User Interface (GUI) including a Graphical User Interface (GUI) element 1101 showing an overview of how product recommendations for a particular user may be assigned according to various categories of product recommendations. Product recommendations may generally be assigned to various categories based on the user's genotype, in which case the graphical user interface element 1101 may be referred to as a "DNA strip" or "DNA product strip. User interface element 1101 may be displayed, for example, by wearable device 100 and/or a personal computing device (e.g., a smartphone).

In this example (and as described above), product recommendations may be divided into three different types, "not recommended," perhaps recommended, "and" recommended. The user interface element 1101 is divided into two "bars" or segments, where each segment corresponds to one of the product recommendation types. The length of the first segment 1102 is proportional to the number of products in the "recommended" category and the length of the second segment 1104 is proportional to the number of products in the "not recommended" category. Generally, the relative lengths of the first and second segments 1102, 1104 are fixed for a particular user, i.e., the portion of the product associated with sedentary sitting and the proportion of the product that is consistently categorized as "recommended" or "not recommended" does not vary with the user's behavior. Thus, the GUI unit 1101 provides each user with a means to determine their respective baseline or "starting point" in terms of different product recommendation categories.

Another part of the products may be referred to as "sedentary related products" (SDP). Product recommendations for SDP are adjusted according to the sedentary behavior of the user. Another GUI element 305 is used to assist the user in monitoring the number of SDPs that he has adjusted. The GUI element 1105 is updated according to the sedentary or sedentary behavior of the user, and thus may be conveniently referred to as a "health bar". Alternatively, it may be referred to as a "DNA health strip" or a "green DNA strip" because it may be used in combination with a "DNA strip". Such designations to emphasize that the GUI element 305 may also help the user realize how their lifestyle reverses or "impairs" some of the advantages that their genetic makeup (DNA) brings. For example, for certain users who are genetically not trending for a risk of obesity but who have a generally inactive lifestyle, product recommendations with more features of users with corresponding genetic predisposition may be presented thereto. The DNA health strip provides a convenient and effective means for users to understand and appreciate how their lifestyle prevents them from "leveraging" their DNA.

In some embodiments, the GUI unit 1105 may be incorporated into the graphical user unit 1101, for example, between the first and second segments 1102, 1104. The DNA product strip 1101 may be updated relatively infrequently, such as when new products are added to the database, or when product recommendations are recalculated based on new biological data provided by the user. In contrast, the DNA health bar 1105 is typically updated multiple times during the course of a day to dynamically indicate the proportion of the product that is recommended to have been affected by the sedentary behavior of the user.

The GUI unit 1105 is divided into: a first sub-segment or "section" 1103A, the length of which is proportional to the number of sedentary related products currently in the "recommended" category according to user behavior; and a second sub-segment or "section" 1103B; the length of which is proportional to the number of sedentary related products currently only in the "perhaps recommended" category depending on the user's behavior. The visual style (e.g., color or shading) of the first section 1103A may be selected to match the first segment 1102 (to indicate that it represents products within the same product recommendation category), while the second section may have a different visual style (to indicate that it represents products within a different product recommendation category). For example, the first segment 1102 and the first section 1103A may both be green, and the second section 1103B may be amber.

The relative lengths of the first and second sections 1103A, 1103B represent the relative proportions of the SDP in which product recommendation adjustments have been made in accordance with the sedentary or sedentary behavior of the user. For example, as more and more "sedentary points" are accumulated by the user, more and more of the SDP's product recommendations are adjusted from "recommended" (or "green") to "perhaps recommended" (or "amber"), and the second section 1103B occupies an increasing proportion of the gui element 1105. The minimum and maximum lengths of the second section 1103B are 0% and 100% of the total length of the GUI unit 1105, respectively. In general, the total length of the graphical user interface element remains unchanged when the length of the second section 1103B is updated, i.e., the amount by which the length of the first section 1103A is shortened (increased) is the same as the amount by which the length of the second section 1103B is increased (decreased).

In some cases, the length of the second section 1103B is reset to 0% at a predetermined time (e.g., midnight). After the reset, the length of the second section 1103B increases as the user's sedentary point accumulates (i.e., its CST increases), but does not decrease as the user engages in physical activity. However, the user may also accumulate points of physical activity (e.g., number of steps) to offset the sedentary points obtained later. In contrast, when the length of the second section 1103B reaches 100%, it does not grow with further sedentary behavior, but the sedentary points can continue to accumulate, thereby necessitating more physical activity to shorten the length of the second section 1103B below 100%.

The GUI unit 1105 may function by allowing a user to set the limit or target (T) of CST_GOAL) (e.g., 4 hours, 6 hours, or 8 hours) is expanded. Wherein the adjusting function of the SDP-related product recommendation may be switched off before the CST of the user reaches said limit value. Alternatively, the limit value may be expressed as a threshold proportion of SDP (e.g., 40%) to allow the corresponding recommendation to be adjusted based on the accumulated CST.

While the DNA health bar 1105 may typically be displayed on a user's personal computing device (e.g., a smartphone or tablet), it may also be displayed on the wearable device 100 if the device itself has a suitable display. In the alternative, wearable device 100 may also alert the user to the effects of his sedentary behavior through some other form of visual indication. For example, the visual indicator (e.g., LED)104 may illuminate in a different color depending on whether the proportion of the product for which the product recommendation has been adjusted exceeds a certain threshold, e.g., green when the threshold is not exceeded; and amber when overtaken. Alternatively, the visual indicator may be illuminated only when the threshold is exceeded. Illumination of the visual indicator may be triggered by one or more inertial sensors that detect a particular motion (e.g., lifting a hand) performed by the user.

FIG. 12 shows the ratio of CST to SDP (referred to herein as R) with the product recommendations adjusted_AMBER) An exemplary relationship or "map" therebetween that takes into account the user-defined CST target. Starting from a starting point where the CST is zero, the SDP ratio increases linearly with increasing CST before reaching the user-defined goal, and thereafter increases non-linearly (although in other examples, the SDP ratio may also increase non-linearly from 0% to 100% throughout). This non-linear relationship may be selected as R_AMBERIncreases at a faster rate with increasing CST until 100% is reached. In the example of FIG. 4, after a user-defined goal is exceeded, R_AMBERIncreases exponentially with increasing CST. This type of non-linear relationship can effectively impose a heavier "penalty" on the user's CST as it increases. The map also enables to target the CST (T) of the user_GOAL) Conversion to an adjusted target ratio (R) for product recommendations_AMBER) The ratio may be used to update the GUI element 1105 to show whether the user remains below their CST target. For example, in some cases, the target scale is explicitly displayed on the GUI unit 1105, e.g., by vertical line 1106.

FIG. 13 shows an example of how a DNA health strip 1105 of a specific user is updated in a 24-hour period. The DNA health strip 1105 may contain an indication of one or both of the amount of physical activity (e.g., number of steps) and the CST for the user. In the figure, text labels are used as the indication, but of course may also take the form of a graphical indication.

The DNA healthy strip was initially 0% (i.e., the second segment 303B was zero in length). In the following step 7: 00-8: 00, the user gets up and moves 5000 steps. During this time period, if the user scans the SDP through the bracelet 100, the product recommendation is not adjusted (e.g., the indicator of the bracelet is green in color). In the next hour, the user accumulates 0.5 hour of Sedentary Time (ST) and 1000 steps of activity (e.g., walking to a station) by traveling on a train. Accordingly, the length of the second section 303B on the health bar 1105 increases to represent a CST of 0.5 hours. However, since the amount of CST is less than the user-set 6 hour goal (represented by the vertical line 306 going to the center of the health bar 1105), there has not been any SDP product recommendation adjusted, i.e., the bracelet still indicates green when scanning for SDP products.

In the following step 9: 00-13: during the 00 time period, the user accumulates CST for 4 hours due to sedentary sitting, causing the length of the second link 303B to increase. At 13: 00-14: during the 00 time period, the user moves more steps, so that the total number of steps is accumulated to be more than 10000 steps (R)_STEP) Thereby causing the CST to decrease and the length of the second link 303B to decrease.

At 14: 00-18: during the 00 time period, the user increases CST to the user-set 6 hour target (T) due to further sedentary sitting_GOAL) This results in the initiation of adjustments to the SDP product recommendations. That is, when the user scans for the SDP, which is only "perhaps recommended" at this time, then the visual indicator of the bracelet 100 will illuminate in amber color to indicate to the user that it has not succeeded in sufficiently reducing his sedentary behavior to be within its selected limit.

At 18: 00-23: during the 00 time period, the user fails to accumulate 10000 steps again, so the number of steps for further activity fails to reduce CST, resulting in further accumulation of CST. However, if the user can move 8000 more steps in that period, it may cause a corresponding decrease in CST. Since CST is greater than T_GOALThus, the length of the second link 303B increases at a more drastic rate as each unit of CST increases.

After the user is asleep, bracelet 100 may, for example, detect that the user is no longer performing any type of activity and stop increasing CST (this operation may also be performed, for example, when the user takes bracelet 100 off). Accordingly, at 23: 00-0: during the 00 time period, the health bar 1105 is no longer updated. However, once midnight has passed, the statistics of health bar 1105 will be reset to enable the user to monitor his sedentary behavior the next day.

Of course, consistent with conventional graphical user interface elements (or "tools"), the overall size or orientation of the graphical user elements 1101, 1105 may be adjusted to match the display of the device or user interface for presenting the user element 301. The relative positions of the individual segments 1102-1104 or sections 1103A, 1103B may be adjusted, for example, as shown in fig. 3, the individual segments may overlap one another adjacent to one another rather than end-to-end. Furthermore, the various segments 302-304 and/or sections 1103A, 1103B may be more easily visually distinguished from one another using colors, shading, or other visual patterns. In some cases, the color scheme used by user interface element 1101 may match the color scheme used when the product code was scanned to indicate a product recommendation to the user on wearable device 100.

Generally, product recommendations can be obtained by the following procedure.

First, several health characteristics that are likely to be associated with potentially adverse health, including, for example, type 2 diabetes, hypertension, high Body Mass Index (BMI), and high cholesterol, are identified. Wherein it may be determined whether the user has any such health characteristics by genetic testing, e.g. based on the detection of Single Nucleotide Polymorphisms (SNPs) in a DNA or RNA sample of the user. In addition, other forms of diagnostic testing may be employed, including testing for other types of biomolecules, or breath testing for determining the composition of the user's microbiome.

Once the "target" health characteristics are determined, the effect of the diet on each health characteristic is initially taken into account. This may be achieved by analyzing each of several different categories relating to the nutritional content of the consumable, including, for example, any of the following: sugar content, calorie content, carbohydrate content, saturated fat content, total fat content, and salt content. For example, it is known that consumption of high-sugar products increases the risk of type 2 diabetes, while a salt-rich diet is associated with hypertension. The relationship between each class and each healthy feature may be quantified by assigning a feature correlation coefficient to each class. Taking hypertension as an example, a larger factor may be assigned to the salt content and fat content, and a smaller factor (or even a zero factor) may be assigned to the calorie content and carbohydrate content.

For each health characteristic determined, a characteristic correlation coefficient may be used to calculate a product score that represents the expected impact based on the nutritional content of the product (i.e., how many grams of salt, saturated fat, etc. the product contains). For example, in calculating a score that represents the adverse effect of a product on hypertension, the salt content coefficient may be multiplied by the number of grams of salt in the product. In mathematical terms, the score for each feature can be determined by taking the scalar product of its feature correlation coefficient vector and the nutritional information vector for each product. Of course, more sophisticated score calculations may also be used, for example, the impact of different nutritional categories on a healthy characteristic may be modeled using a determined dose response curve for that characteristic. The score may also take into account other factors such as product type (e.g., candy, cookies, breakfast cereal, etc.) and typical serving size of the product. Such other factors may be used to adjust the score for differences in consumption patterns of different products. For example, if a product is determined to be generally consumed as a "snack," its score may be reduced, thereby making recommendations for that product more positive than would otherwise be given.

After the score for each product is calculated, the products are ranked in descending order of their score for a particular health feature. A portion of the products are then selected by selecting products that rank above a threshold rank. For example, the top 50%, 30%, or 10% ranked products may be selected based on their health feature scores. Such products may be assigned "negative" recommendations, such as "not recommended". As such, when the bracelet 100 reads the barcode of the product, the indicator (LED)106 turns red (for example). The remaining products are assigned other categories such as "recommended" (the user will be prompted in green). Thus, each product assigns different recommendations depending on its ranking for a particular feature. The threshold ranking value (or "cutoff value") used to assign recommendations varies from user to user and is determined by how susceptible the user is to the particular feature (e.g., as determined by genetic testing).

The total product recommendation for each product is determined by combining the product recommendations for each health feature. This may be accomplished in a number of ways, such as determining a recommendation for the product as "not recommended" if there is a recommendation for any feature as "not recommended" or if there is a recommendation for more than one feature as "not recommended". Further, for each product, a flag indicating whether the product recommendation should be adjusted may be set or not set according to the same or similar rule as above. For example, after generating a product recommendation, each product code may be associated with a "tag". One value of the tag (e.g., 0) may be used to indicate non-SDP, while the other value may be used to indicate a product code associated with SDP. In some cases, CST may be used to define a "start-up threshold" value for the tag, so that all products with tag values above the threshold are adjusted.

The portion of the product whose recommendations may be adjusted based on the sedentary behavior of the user may be selected in a variety of ways. Typically, no adjustments are made to all types of food products, as this may make "healthy" products such as vegetables "perhaps recommended" due to the sedentary behavior of the user. The portion of products may be selected based on nutritional data associated with each product. In one example, for products classified as "recommended," the products are ranked according to their calorie content, and products with the top 50%, 30%, or 10% of the calorie content among the products are selected as products for adjustment. In some embodiments, the user may adjust the cutoff percentage used when selecting products in this manner to increase or decrease the number of products being adjusted.

The quantity of products whose associated product recommendations are adjusted (among the above-mentioned portion of products) is determined in accordance with the CST described above in connection with FIG. 4.

In general, the above threshold rankings and the rules for determining total product recommendations are balanced in two ways: allowing the user to select the product that the user wishes to consume; users are discouraged from consuming at least a portion of the product that is most likely to have an adverse effect on their health. The advantage of this balance is that the user can be "gently persuaded" to make a choice that is more favorable for him in the long term. Adjusting product recommendations based on a user's sedentary and/or other sedentary behavior may further improve the additional effect of "gentleness persuasion" -for example, a user may find that a certain originally "recommended" product becomes a "perhaps recommended" ("discretionary") product after a long period of sedentary time within a certain day. Not adjusting the product to "not recommended" may result in the following particularly significant advantages: in certain situations where the user is not in control of himself, for example, where the user has to sit for a long distance, no overly severe penalty is imposed on the user.

The steps of the method of alerting a user to the negative effects of sedentary behavior are as follows:

step 1: the product code of the consumable and the data representing or available to obtain the corresponding product recommendation are stored in a memory of the computing device.

Step 2: motion data is obtained from one or more inertial sensors worn by the user.

And step 3: the behavior of the user is monitored by using the motion data to determine a period of time that the user is in a sedentary position or other sedentary state.

And 4, step 4: and selecting a product code for adjusting the corresponding product recommendation from at least one part of the product codes according to the behaviors of the user.

And 5: the visual indicator is controlled to provide a visual indication based on the number of product codes selected.

FIG. 14 illustrates an exemplary graphical user interface 701 containing a DNA strip 1101. FIG. 15 illustrates an exemplary graphical user interface 801 including a health bar 1105. Each of the GUIs can be rendered.

An example system includes a personal computing device (such as a smartphone, a smartwatch, a tablet computer, or a desktop device) and a wearable device 100. The personal computing device includes: a transceiver for exchanging data with the wearable device 100 (e.g., through a wired or wireless connection); a memory for storing data; and a processor for processing data received from the wearable device 100 and controlling the display to display the DNA strip 301 and/or the health strip 1105. In use, wearable device 100 may: sending the motion data to the personal computing device for the personal computing device to use the motion data to monitor the user's behavior to determine a time period during which the user is in a sedentary position or other sedentary state; and selecting a product code to be adjusted according to the behavior of the user. Alternatively, the wearable device 100 itself may implement the monitoring and selection, in which case the wearable device 100 sends data representing the selected product code to the personal computing device. The processor updates the display content, for example by updating the health bar 1105, to provide a visual indication that depends on the number of products selected.

The memory may store product codes and data representing or available to obtain recommendations for respective products. The product code and data may be downloaded into the wearable device 100, for example, after the product code and/or product recommendation are updated.

While the above description focuses on sedentary or other sedentary oligodynamic behavior, the various aspects described above may further (or alternatively) be equally applicable to other aspects of the user's lifestyle and/or the user's environment. For example, as described above, in some cases, the wearable device may include an ultraviolet sensor that may be used to monitor the user's ultraviolet exposure (of course, body worn sensors that are not part of the wearable device 100 may also be used). In this case, the user's product recommendations may be adjusted based on their behavior with respect to uv exposure (e.g., the time of exposure to direct sunlight). Subsequently, the wearable device 100 or personal computing device can be used to provide a visual indication (e.g., via the DNA health strip 305) of the recommended number of products to be adjusted depending on the length of time the ultraviolet light is exposed. In this manner, the user may be alerted to the harmful effects that ultraviolet light exposure may have (e.g., by recommending a product that will increase vitamin E levels). Similarly, the system may also be configured to provide feedback to the user (in terms of adjusted recommended quantities of product) regarding the harmful effects that may exist in their environmental composition based on data obtained from body-worn air quality or pollutant (e.g., NOx or particulate matter) sensors.

In another example, product recommendations may be adjusted based on the length of time and/or intensity of a particular physical activity (or physiological function) such as running by the user (based on data obtained by one or more sensors such as a pedometer). For example, some users may be genotyped as likely to have lower bone density and therefore may require appropriate adjustments to certain product recommendations to favor certain footwear that can mitigate the effects of high impact activity on the user's skeletal system. By providing a visual indication to the user of this adjustment, the user may be alerted to the need to take remedial action, such as performing a lower impact activity.

In some embodiments, a plurality of DNA health bars 1105 (or other forms of visual indicators) may be provided (e.g., in the form of "tiles" in a user interface) to enable a user to track the relative degree of contribution of various different user lifestyles and user environmental factors to the number of product recommendations to be adjusted. For example, when a user reduces sedentary time by walking, if he is walking in direct sunlight, or is walking in a contaminated area, the number of adjusted product recommendations may only slightly decrease (or even increase). Thus, by alerting the user to the problem with multiple visual indications, the user may be enabled to more appropriately adjust his or her behavior.

It will be appreciated by those skilled in the art that various modifications can be made to the above-described embodiments without departing from the scope of the invention. For example, although the above described principle embodiment is described in the form of a wearable bracelet, the above described system may also take any other suitable form, such as a graspable barrel, key ring, pendant or smart phone, or any combination of such forms. In addition, it should be noted that the data stored in the system may be derived from biological information obtained by analyzing user-provided biological samples as well as other user-provided samples. The group of users may be members of the same family. Thus, the data store contains a set of public data that can be used to provide the best recommendations for all family members.

Claims (43)

1. A wearable device, comprising:

a memory for storing product codes for consumables, topical application products and/or body worn products and storing data representing respective product recommendations or data representing product recommendations available for acquisition;

a product code reader for reading a product code from a product;

one or more inertial sensors for obtaining motion data of a wearer of the device;

a visual indicator responsive to the read product code for providing a visual indication of a product recommendation using the data stored in the memory; and

one or more processors to process the motion data to determine a period of time that the wearer is in a sedentary position or other sedentary state; analyzing the occurrence time and the duration of the time period; and adjusting the recommendation of at least a portion of the product code accordingly, wherein product recommendations vary according to the determined time period.

2. The wearable device of claim 1, wherein the one or more processors are configured to adjust recommendations based on the processor determining the wearer's behavior over a preset length of time, such as a day.

3. The wearable device according to claim 1 or 2, wherein the one or more processors are configured to determine and maintain the length of time the wearer is in the sedentary position or other sedentary state for longer than a length threshold number of time periods, such as 10, 30 or 60 minutes.

4. The wearable device according to any of the preceding claims, wherein the one or more processors are configured to ignore interrupt times for which the user is not in a sedentary position or other sedentary state that are less than a predetermined time when determining the time period.

5. The wearable device according to any of the previous claims, wherein the memory stores nutritional data for each product, and the processor is configured to select a portion of the product code based on the nutritional data.

6. Wearable device according to any of the previous claims, where each product recommendation has a first state of-recommended and a second state of-not recommended, the adjustment being such that the product recommendation changes between said first and second state.

7. Wearable device according to any of claims 1-5, characterized in that the product recommendation has a first state of-recommended, -a second state of perhaps recommended, and-a third state of not recommended, the adjustment being such that the product recommendation changes between the first and second state.

8. Wearable device according to any of the previous claims, characterized in that it is a wrist worn wearable device.

9. The wearable device according to any of the preceding claims, wherein the one or more inertial sensors comprise an accelerometer and/or a gyroscope.

10. Wearable device according to any of the previous claims, characterized in that the product code reader is used to read one-dimensional or two-dimensional bar codes.

11. Wearable device according to any of the previous claims, characterized in that the visual indicator provides an indication of the product recommendation by illumination of different colors, such as red and green, or red, green and amber.

12. The wearable device according to any of the preceding claims, wherein the one or more processors are configured to select a product code from at least a portion of the product codes to be adjusted for a respective product recommendation based on the user's behavior; and the or another visual indicator is controllable by the one or more processors to provide another visual indication, the another visual indication being dependent on the number of product codes selected.

13. A method of improving the health of a wearer of a wearable device by providing recommendations on consumable items to the wearer, the method comprising:

storing a product code for a product and data representing a corresponding product recommendation or data representing product recommendations available in a memory of the wearable device;

obtaining motion data of the wearer from one or more inertial sensors of the wearable device;

processing the motion data with a processor of the wearable device to determine a period of time that the wearer is in a sedentary posture or other sedentary oligodynamic state;

analyzing, with the processor, the time of occurrence and the duration of the time period;

reading a product code from a product with a product code reader of the wearable device;

using data stored in the memory in response to the read product code to obtain a product recommendation;

adjusting the product recommendation based on the sedentary behavior or other sedentary oligodynamic behavior of the wearer determined by the processor; and

providing, with a visual indicator of the wearable device, an adjusted visual indication of the product recommendation.

14. The method of claim 13, further comprising: a portion of the product code that is adjusted in relation to the product recommendation is selected.

15. The method of claim 14, wherein selecting a portion of the product code for which product recommendations are adjusted comprises: the product code is selected based on nutritional information about the product.

16. The method of claim 15, wherein selecting a product code based on the product-related nutritional information comprises: sorting the products according to the corresponding nutritional information of each product; and selecting the portion of product codes by selecting products ranked above a predetermined rank.

17. The method according to claim 15 or 16, wherein the nutritional information comprises one or more of calorie content, sugar content, carbohydrate content, saturated and/or unsaturated fat content, and salt content.

18. The method according to any one of claims 13 to 17, wherein the data is derived from personalized data derived from personalized biological information obtained by analyzing a biological sample provided by a wearer of the device.

19. A method of benefiting a user's health by providing product recommendations in consumable items, the method comprising:

identifying one or more health characteristics of the user;

for each of a plurality of consumables, topically applied products, and/or body worn products, calculating a score representing a degree of influence of the product on each of the health characteristics, each score based at least in part on nutritional information for the product;

assigning a product recommendation to each product based on the score;

monitoring the user's behavior by determining a period of time that the user is in a sedentary position or other sedentary oligodynamic state;

adjusting a product recommendation for at least a portion of the product based on the user's behavior; and

providing the adjusted product recommendation to the user via a visual indicator.

20. The method of claim 19, wherein the product recommendation for each product is stored in a database disposed in a memory of a wearable device, the memory further storing data indicative of whether the product is part of the product.

21. The method of claim 20, wherein the wearable device comprises one or more inertial sensors for determining a period of time that the user is in a sedentary or other sedentary posture.

22. A monitoring system for alerting a user to adverse effects caused by a user's environment and/or lifestyle, the monitoring system comprising:

a memory storing product codes for consumables, topical application products and/or body worn products and data representing corresponding product recommendations or data representing available corresponding product recommendations;

one or more processors configured to:

monitoring the behaviour of the user using data representative of one or more physiological and/or biochemical functions of the user or representative of the user's environment, said data being obtained by one or more sensors worn by the user; and

selecting a product code recommended by a corresponding product to be adjusted from at least one part of the product codes according to the behavior of the user; and

a visual indicator controllable by the one or more processors to provide a visual indication according to the number of product codes selected.

23. A monitoring system for alerting a user to the negative effects of sedentary behavior, comprising:

a memory storing product codes for consumables, topical application products and/or body worn products and data representing corresponding product recommendations or data representing available corresponding product recommendations;

one or more processors configured to:

monitoring the user's behavior by identifying a period of time that the user is in a sedentary posture or other sedentary oligodynamic state using motion data obtained from one or more inertial sensors worn by the user; and

selecting a product code recommended by a corresponding product to be adjusted from at least one part of the product codes according to the behavior of the user; and

a visual indicator controllable by the one or more processors to provide a visual indication according to the number of product codes selected.

24. The monitoring system of claim 23, wherein the one or more processors are configured to determine a count of a number of time segments during which the user is in the sedentary position longer than a time threshold and increase the number of selected product codes as the count increases.

25. The monitoring system of claim 24, wherein the one or more processors are configured to increase the number of selected product codes by successively increasing amounts as the count increases.

26. The monitoring system of claim 24 or 25, wherein the one or more processors are configured to determine the amount of physical activity performed by the user from the received athletic data and to decrement the count for the period of time based on the amount of physical activity performed by the user.

27. The monitoring system of claim 26, wherein the one or more processors are configured to determine a number of steps of the user activity based on the athletic data.

28. The monitoring system of any one of claims 24 to 27, wherein the one or more processors are configured to implement the adjustment to the product recommendation only if the count exceeds a predetermined cutoff value.

29. The monitoring system of claim 28, further comprising a user interface for receiving user input, the processor being configured to adjust the cutoff value based on the user input.

30. A monitoring system according to any one of claims 23 to 29, wherein each product recommendation has a first state of-recommended and a second state of-not recommended, the adjustment being such that the product recommendation changes between said first and second states.

31. A monitoring system according to any one of claims 23-29, wherein said product recommendation has a first state of-recommended, -a second state of perhaps recommended, and-a third state of not recommended, said adjustment being such that the product recommendation changes between said first and second states.

32. The monitoring system according to any one of claims 23 to 31, wherein the visual indicator is for displaying a graphical element having a length or area representing a number or proportion of selected product codes.

33. A monitoring system according to any of claims 23 to 32, the system being configured to: receiving a product code; retrieving or deriving a product recommendation for the consumable product associated with the product code; and adjusting the product recommendation if and only if the product code is one of the selected product codes.

34. The monitoring system of claim 33, further comprising a product code reader for reading a product code from a product.

35. The monitoring system of any one of claims 23 to 34, wherein the processor is configured to select a product code to be adjusted for a respective product recommendation based at least in part on nutritional information about the product.

36. The monitoring system of any one of claims 23 to 35, wherein the one or more processors are configured to: sorting the products according to the corresponding nutritional information of each product; and selecting the product code by selecting a product ranked above a predetermined rank.

37. The monitoring system of claim 36, wherein the nutritional information includes one or more of calorie content, sugar content, carbohydrate content, saturated and/or unsaturated fat content, and salt content.

38. The monitoring system of any one of claims 23 to 37, wherein the data is derived from personalized biological information obtained by analysis of a biological sample provided by the user.

39. The monitoring system of any one of claims 23 to 38, wherein the visual indicator is a display of a personal computing device such as a smartphone or a smartwatch.

40. A wearable monitoring device that alerts a user to the negative effects of sedentary behavior, comprising:

a memory storing product codes for consumables, topical application products and/or body worn products and data representing corresponding product recommendations or data representing available corresponding product recommendations;

one or more inertial sensors for obtaining motion data of a wearer of the device;

one or more processors configured to:

monitoring the behavior of the user by determining a period of time that the user is in a sedentary position or other sedentary oligodynamic state using the motion data; and

selecting a product code recommended by a corresponding product to be adjusted from at least one part of the product codes according to the behavior of the user; and

a visual indicator controllable by the one or more processors to provide a visual indication according to the number of product codes selected.

41. A method of alerting a user to the consequences of sedentary behavior, the method being implemented by one or more computing devices and comprising:

storing in a memory of at least one of the computing devices a product code of a consumable and data representing a respective product recommendation or data representing respective product recommendations available for acquisition;

obtaining motion data from one or more inertial sensors worn by the user;

monitoring the user's behavior by determining a period of time that the user is in a sedentary position or other sedentary oligodynamic state using the motion data;

selecting a product code to be adjusted for a corresponding product recommendation from at least a portion of the product codes according to the behavior of the user; and

the visual indicator is controlled to provide a visual indication in accordance with the number of product codes selected.

42. The method of claim 41, wherein the selected product code is selected from a portion of product codes for which corresponding product recommendations can be adjusted, and further comprising: the portion of the product code is selected based on nutritional information about the product.

43. The method of claim 42, further comprising: ranking the products according to the respective nutritional information for each product, wherein selecting the product code comprises: products ranked above a predetermined rank are selected.

Images