WO2009156936A2 - System and method for determining a personal health related risk - Google Patents

System and method for determining a personal health related risk Download PDF

Info

Publication number
WO2009156936A2
WO2009156936A2 PCT/IB2009/052663 IB2009052663W WO2009156936A2 WO 2009156936 A2 WO2009156936 A2 WO 2009156936A2 IB 2009052663 W IB2009052663 W IB 2009052663W WO 2009156936 A2 WO2009156936 A2 WO 2009156936A2
Authority
WO
WIPO (PCT)
Prior art keywords
sensor data
risk
system
data
personal health
Prior art date
Application number
PCT/IB2009/052663
Other languages
French (fr)
Other versions
WO2009156936A3 (en
Inventor
Calina Ciuhu
Warner R. T. Ten Kate
Jose Fernandez
Original Assignee
Koninklijke Philips Electronics N.V.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to EP08159152 priority Critical
Priority to EP08159152.1 priority
Application filed by Koninklijke Philips Electronics N.V. filed Critical Koninklijke Philips Electronics N.V.
Publication of WO2009156936A2 publication Critical patent/WO2009156936A2/en
Publication of WO2009156936A3 publication Critical patent/WO2009156936A3/en

Links

Classifications

    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F19/00Digital computing or data processing equipment or methods, specially adapted for specific applications
    • G06F19/30Medical informatics, i.e. computer-based analysis or dissemination of patient or disease data
    • G06F19/32Medical data management, e.g. systems or protocols for archival or communication of medical images, computerised patient records or computerised general medical references
    • G06F19/321Management of medical image data, e.g. communication or archiving systems such as picture archiving and communication systems [PACS] or related medical protocols such as digital imaging and communications in medicine protocol [DICOM]; Editing of medical image data, e.g. adding diagnosis information
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients

Abstract

A system (10) is provided for determining a personal health related risk (25, 312). The system (10) comprises a sensor input (11) for receiving sensor data related to a health parameter of a patient, a sensor data analysis unit (13) for analyzing and classifying the received sensor data, a context input (12) for receiving contextual data, and a reasoning unit (14) for performing a statistical analysis of the contextual data and of an output of the sensor data analysis unit (13), and for, based on the statistical analysis, determining the personal health related risk (25, 312). The personal health risk (25, 312) may, e.g., be a melanoma risk (25) or a risk of falling (312). The statistical analysis may, e.g., use Bayesian statistics.

Description

System and method for determining a personal health related risk

FIELD OF THE INVENTION

This invention relates to a system for determining a personal health related risk, the system comprising a sensor input for receiving sensor data related to a health parameter of a patient and a sensor data analysis unit for analyzing and classifying the received sensor data in order to determine the personal health related risk.

This invention further relates to a method of determining a personal health related risk, the method comprising the steps of receiving sensor data related to a health parameter of a patient and analyzing and classifying the received sensor data in order to determine the personal health related risk. This invention also relates to a computer program product for performing said method.

BACKGROUND OF THE INVENTION

Decision support systems (DSS) are important tools in the medical field, which help reducing the workload in the hospital by providing means for disease identification. Existing automatic sensor data analysis tools may assist a physician in determining whether a patient has a certain disease or other health risk. Besides using the sensor data analysis tools, the physician also uses his knowledge about the life and health of the patient and his own experience as a physician to determine the risk of the patient having a particular disease or health risk. Consequently, the risk assessment can only be performed by or with help from the physician. This makes the risk assessment costly, time consuming and unpractical. Therefore, there is a need for an easier way to obtain an accurate assessment of a personal health related risk.

Another example of a system for determining a personal health related risk is a fall prevention system. Sensors register movements of the patient and the registered movements are analyzed in order to determine whether the patient is in balance or is expected to fall. Falls and fall-related injuries are among the most serious and common medical problems experienced by older adults. Nearly one-third of older persons fall each year, and half of them fall more than once. Because of underlying osteoporosis and decreased mobility and reflexes, falls often result in hip fractures and other injuries, head injuries and even death in older adults. The analyzed sensor data, e.g. in the form of a balance score based on the movement data, may trigger a warning for the patient and/or other persons being close to the patient. Unfortunately, sensor data alone is often not enough to make an accurate estimation of a person's health risk. A particular sensor value may indicate a high risk for one person, while for another person the same value may indicate an intermediate or low risk.

Although the proposed invention may be used for determining health related risks for many different diseases and health risks, the remainder of the description will focus on fall prevention and malignant melanoma. Malignant melanoma is the most aggressive and deadliest form of skin cancer. The estimated number of new cases and deaths in the USA in 2008 is 62,000 and 8,500, respectively. When detected at an early stage, melanoma can be treated effectively at a relatively low cost. This results in a high 5 -year survival rate of 95%. Progression of the disease in the late stage is associated with poor survival rate of 13%. Therefore earlier detection of melanoma is essential and still one of the most challenging problems in dermatology.

Screening is nevertheless difficult, partially because it would overload the specialist or because people are reluctant to visit a specialist. Currently, the expert makes the diagnosis not only by visual examination of the nevi, but also by integrating this information with the patient profile (such as family history, frequency of holidays or sun exposure, etc...). To enable large-scale screening, this procedure has to be brought outside the hospital and needs to be accessible. At this moment, there is no easy-to-access tool for skin inspection with respect to melanoma. As a result, the general population does not have easy access to an accurate diagnostic tool.

OBJECT OF THE INVENTION

It is an object of the invention to provide a system and method for accurately determining a personal health related risk, requiring little interference by a physician or other person.

SUMMARY OF THE INVENTION

According to a first aspect of the invention, this object is achieved by providing a system for determining a personal health related risk, the system comprising a sensor input, a sensor data analysis unit, a context input and a reasoning unit. The sensor input is provided for receiving sensor data related to a health parameter of a patient. The sensor data analysis unit is provided for analyzing and classifying the received sensor data. The context input is provided for receiving contextual data. The reasoning unit is provided for performing a statistical analysis of the contextual data and of an output of the sensor data analysis unit and for, based on the statistical analysis, determining the personal health related risk.

By combining sensor data analysis tools with contextual knowledge, the accuracy can be improved. The sensor data analysis tool classifies the sensor data based on health parameters present in the sensor data, while the reasoning unit refines the risk estimate by incorporating knowledge about the context of the sensor data or patient. An important aspect of the invention is that the output of the sensor data analysis unit is used as input for the reasoning unit. The sensor data analysis results and the contextual data are used for a combined statistical analysis of the available information. This combined statistical analysis provides much more accurate results than sensor data classification alone. It is an advantage of the system according to the invention that it does not rely on the experience and intuition of the physician and is therefore arranged for providing an objectively measurable personal health related risk. The system according to the invention also reduces the time required for making an analysis of a particular health related risk. Consequently, the work load of the physician may be reduced. It is another advantage of the system according to the invention that it significantly improves the usefulness of sensor-based health advisers by taking advantage of contextual data.

The personal health related risk may, e.g., be a melanoma risk, the sensor data comprising an image of a skin lesion of the patient. The personal health risk may also be a risk of falling, the sensor data comprising movement data of the patient. Also other types of health related risks and sensor data fall within the scope of the invention.

In an embodiment of the system according to the invention, the output of the sensor data analysis unit comprises a single classification of the personal health related risk based on the received image. This classification is then used as an input parameter for the statistical analysis. In addition to the classification, the output may comprise an accuracy of the classification. If the accuracy of the classification is high, the contextual data may have little influence on the outcome of the statistical analysis. If the accuracy of the classification is low, the contextual data will be more important.

Alternatively, the sensor data analysis unit is arranged for extracting multiple parameters from the received sensor data and the output of the sensor data analysis unit comprises the multiple features. For example, image analysis tools are used for extracting different features from an image. The presence, absence or relative or absolute value of each extracted feature may influence the classification of the image. Instead of only using the classification as input for the reasoning unit, the system may also use the separate extracted features as input for the statistical analysis. Also each or some extracted features may have a corresponding confidence level.

According to a second aspect of the invention, a method is provided comprising the steps of receiving sensor data related to a health parameter of a patient, analyzing and classifying the received sensor data, receiving contextual data, performing a statistical analysis of the contextual data and the analyzed and classified sensor data, and based on the statistical analysis, determining the personal health related risk.

According to a third aspect of the invention, a computer program product is provided.

These and other aspects of the invention are apparent from and will be elucidated with reference to the embodiments described hereinafter.

BRIEF DESCRIPTION OF THE DRAWINGS

In the drawings:

Fig. 1 shows a block diagram of a system according to the invention. Fig. 2 shows a statistical model for use in a system for determining a melanoma risk, and

Fig. 3 shows a statistical model for use in a system for determining a risk of falling.

DETAILED DESCRIPTION OF THE INVENTION

Figure 1 shows a block diagram of a system 10 according to the invention. The system 10 comprises two inputs 11, 12. A sensor input 11 is provided for receiving sensor data related to a health parameter of a patient. The sensor data may, e.g., be image data representing pictures of human skin, an X-ray image of bones and joints or movement data representing the movements of a patient. The movement data can represent several physical quantities, such as accelerometer data, magnetometer data, gyroscope data. In fact, the sensor data may be any kind of measurable data which may be used as a measure of some health related parameter. Examples include physiologic data and other vital body signs. Of course, the device for obtaining the sensor data may be part of the system 10. Alternatively, the sensor data are provided via a data network or information carrier. The data may also be entered manually via user interface elements, such as a keyboard or a mouse. In sensor data analysis unit 13, the received sensor data is analyzed and classified.

The analysis performed by analysis unit 13 may, e.g., result in a score or classification representing a health risk, based on the sensor data only. An image analysis tool 13 for classifying a melanoma risk may, e.g., classify an input image of a skin lesion as either malign or benign. Alternatively, a score between 0 and 100 may be provided, 100 corresponding to a high risk of the skin lesion being malign and 0 corresponding to a high probability of the skin lesion being benign. The output value or classification of the sensor data analysis unit 13 may be obtained by first determining one or more intermediate values or parameters. Instead of providing only one output value or classification, the sensor data analysis unit 13 may provide these intermediate values or parameters as output for use in the statistical analysis. In addition to the output values and classifications, a confidence level or accuracy measure for the output values may be provided. This can be a single (scalar) value, such as a variance value, but also a vector or multi-dimensional matrix, such as a covariance matrix, higher order moments, or other forms of characterizing the accuracy. Of course highly accurate values or classifications are more important than less accurate ones. The accuracy or confidence level of the output of the sensor data analysis tool 13 may influence the definitive personal health risk as determined by the system 10 according to the invention. When the sensor data analysis tool 13 is an image analysis tool 13 for analyzing images of skin lesions, the analysis may be based on the so-called ABCD features, which are associated with Asymmetry, Borders, Color and Differential structures, respectively. For each of the features a value is calculated. Depending on the specific feature, the calculated risk results as a weighted combination of the A, B, C, and D values. The feature extraction and contribution varies among different experts and their level of expertise. A trained classifier 13 based on pattern recognition can simulate this process and can even include various other features which, because they lack interpretability, are not used by experts. The classifier 13 is provided with several pictures that are labeled as being benign or malign, and the training algorithm optimizes the feature weighting for optimal classification of future images. Besides automatic feature selection, the classifier 13 can provide corresponding contribution weights, which optimize the image based disease identification. When the system 10 is used for determining a risk of falling, the sensor data may come from movement detectors worn by the user or installed in the area where the user is staying. The sensor data analysis tool 13 may use this data to determine a balance score or other value representing a risk of falling, purely based on measured movement data. For example, from the sensor data the steps taken by the user can be estimated, from which subsequently their variance is observed. A high variance indicates an unstable situation. Other measures are conceivable as well. A context input 12 is provided for receiving context information. The type of context data received depends on the type of personal health risk that is determined. When determining a melanoma risk, dermatologists do not base their diagnosis solely on the analysis of the skin lesion. Patient related context is also of relevance in a diagnosis. The same lesion could be diagnosed as highly dangerous if it is on the leg of an Irish female and benign if the bearer is an Italian male. This relates to the inherent overlap in feature distribution of malignant and benign lesions. The dermatologist distinguishes different populations characterized by factors like age, gender, sun-burn history, hair color, number of nevi, and personal as well as family disease history. The risk population the patient belongs to is taken into account in the diagnosis. If the system 10 is used for determining a melanoma risk, the context data may thus comprise, e.g., a skin type, an age, a gender, a sun-burn history, a hair color, a number of nevi, a personal or family disease history or a part of the body comprising the skin lesion. If the system 10 is used for determining a risk of falling, the context data may comprise, e.g., an age, a history of falling, a chronic disorder, medication use or a general health condition. Additionally, external, environmental variables, such as ground surface regularity or lighting condition influence the risk of falling and may thus be provided to the context input 12.

Received context data may be stored in a storage means 15, such that it can be used later on for an additional or another risk measurement. Together with or separate from the contextual data, epidemiological data may be received/stored. For example, men have a higher risk of melanoma in their trunk than women. On average, 25% of all melanoma in men occurs in the legs. Melanoma is very rare for persons under 10 years of age, but more than half of all cases of melanoma occur with people over 60 years of age. Such epidemiological information, in combination with the other more personal contextual data, can be used for making a statistical analysis of the melanoma risk. Preferably, the epidemiological data is received and updated on a regular basis (e.g. monthly, yearly), e.g., via the Internet or via an update CD or DVD. The more personal contextual data may, e.g., be updated each time the data changes or each time a person uses the system. The database can be remote from the system, e.g. being accessible over the Internet. The maintenance (updating) can be performed by a third party, preferably knowledgeable about the respective health risk, its etiology and its epidemiology.

The received contextual data and the output of the sensor data analysis unit 13 are both sent to the reasoning unit 14 for performing a statistical analysis of the contextual data and the output of the sensor data analysis unit 13. Based on the statistical analysis, the personal health risk is determined.

The reasoning unit 14 incorporates algorithms that infer conclusions when provided with observations, by using some formal model of the domain. A typical example is diagnosis, where the formal model consists of a description of a disease's etiology and the observations are symptoms, complaints of the patient, and possibly outcomes of tests. The reasoning unit 14 infers possible disorders that can cause (explain) the observed symptoms. The formal model can be symbolic, which means that its scheme mainly holds causal dependencies, such as in a rule-based scheme (IfA, then B, where A and B are the symbols that represent disorders, symptoms, anatomy, etc.). The formal model can also be probabilistic, in which case mainly the joint-probabilities between the different entities are given. Such a scheme allows computing the most likely disorder that can have caused the observed symptom. The Bayesian Network (BN) is the prototypical example of a statistical model that may be used for the reasoning unit 14 according to the invention.

Because the formal model consists of a description of a disease's etiology and observations, reasoning systems reflect human knowledge. Experts of the domain have defined the concepts and their relationships that together span the model. However, it is also possible to create a model through learning from data that represent the domain. In particular, this is possible with BN. Hence, BN can both be used to build a classifier as well as to build a reasoning system 14. According to the invention, the reasoning unit 14 combines classification based on sensor data with reasoning based on human expert knowledge, so that a personal health risk is determined, which risk determination is improved over that of either classifier or reasoner alone.

Figure 2 shows a statistical model for use in a system 10 for determining a melanoma risk 25. The model shown in Figure 2 uses four risk factors: skin type 21, age 22, gender 23 and part of the body 24. The contextual data defines these risk factors for the patient. For example, a skin type classification may distinguish six different classes (I, II, II, IV, V, VI) of skin type from very white to very dark skin. Melanoma risk 25 of a person with skin type I is about 10 times higher than for a person with skin type VI. A skin lesion found in an input image is more likely to be a benign nevus for a person with skin type VI than for a person with skin type I.

In a similar way, the melanoma risk 25 is influenced by age 22, gender 23 and part of body 24. For each factor the population may be divided into different classes or some functions may define the risks for different populations. All four factors shown in Figure 2 have a direct influence on the melanoma risk 25 of a person. Additionally, the risk factor gender 23 also influences the other risk factor part of body 24. For men, the melanoma risk is relatively high if the lesion is found in the head, neck or trunk. Women are more likely to show melanoma in the legs. According to the invention, the statistical model does not only comprise contextual risk factors 21, 22, 23, 24, but also the output 26 of the sensor data analysis unit 13. If, e.g., the sensor data analysis unit 13 determines that a skin lesion is malign, this is used as input data 26 for the statistical model. In, e.g., a Bayesian network, the general TP (true positive), FP (false positive), TN (true negative) and/or FN (false negative) rate of the sensor data analysis unit 13 may be used to define the conditional probabilities between the sensor data analysis unit output 26 and the melanoma risk 25. Alternatively or additionally, an accuracy calculated by the sensor data analysis unit 13 may be used to adapt or replace the static TP, FP, TN and FN values at input 26 for the statistical model. Instead of only using an ultimate output of the sensor data analysis unit 13 as input 26 for the statistical analysis, the sensor data analysis may also provide a set of parameters determined from the received sensor data and/or multiple classification results as determined from different measurements/ images. If the data analysis results in multiple output parameters, each parameter may be represented as a separate node in the statistical model. Each node may be related to one or more other nodes. Nodes related to contextual data may be related to nodes related to sensor data output.

Figure 3 shows a statistical model for use in a system for determining a risk of falling 312. In this event, the output of the sensor data analysis unit 13 may be a balance score 307 obtained from, e.g., movement detection sensors. Other sensors may measure lighting conditions 310 or type of surface 309 and the data from these sensors may also be processed for use in the statistical model. Bad lighting and uneven surfaces will increase the risk of falling 312. Preferably, the memory 15 comprises data describing a relation between lighting conditions 310/ surface type 309 and risk of falling 312. The contextual data may include some personal data of the patient. Important patient related risk factors are age 301, fall history 302, fear of falling 303, chronic disorders 304, medication use 305 or a general health condition 306. In this example, the general health condition 306 is influenced by the nodes 301-305 and the general health condition 306 influences the balance score 307 determined by the sensor data analysis unit 13. Further nodes in the statistical model may represent the number of tasks 313 the user is performing, for example walking and talking, sensor readings 314 that represent an activity level of the user and the amount of variation 315 in the user's gait. The last mentioned free nodes 313, 314, 315 may together form the stability state 316 the user is in.

It will be appreciated that the invention also extends to computer programs, particularly computer programs on or in a carrier, adapted for putting the invention into practice. The program may be in the form of a source code, an object code, a code intermediate source and object code such as in partially compiled form, or in any other form suitable for use in the implementation of the method according to the invention. It will also be appreciated that such a program may have many different architectural designs. For example, a program code implementing the functionality of the method or system according to the invention may be subdivided into one or more subroutines. Many different ways to distribute the functionality among these subroutines will be apparent to the skilled person. The subroutines may be stored together in one executable file to form a self-contained program. Such an executable file may comprise computer executable instructions, for example processor instructions and/or interpreter instructions (e.g. Java interpreter instructions). Alternatively, one or more or all of the subroutines may be stored in at least one external library file and linked with a main program either statically or dynamically, e.g. at run-time. The main program contains at least one call to at least one of the subroutines. Also, the subroutines may comprise function calls to each other. An embodiment relating to a computer program product comprises computer executable instructions corresponding to each of the processing steps of at least one of the methods set forth. These instructions may be subdivided into subroutines and/or be stored in one or more files that may be linked statically or dynamically. Another embodiment relating to a computer program product comprises computer executable instructions corresponding to each of the means of at least one of the systems and/or products set forth. These instructions may be subdivided into subroutines and/or be stored in one or more files that may be linked statically or dynamically.

The carrier of a computer program may be any entity or device capable of carrying the program. For example, the carrier may include a storage medium, such as a ROM, for example a CD ROM or a semiconductor ROM, or a magnetic recording medium, for example a floppy disk or hard disk. Further, the carrier may be a transmissible carrier such as an electrical or optical signal, which may be conveyed via electrical or optical cable or by radio or other means. When the program is embodied in such a signal, the carrier may be constituted by such a cable or other device or means. Alternatively, the carrier may be an integrated circuit in which the program is embedded, the integrated circuit being adapted for performing, or for use in the performance of, the relevant method.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. Use of the verb "comprise" and its conjugations does not exclude the presence of elements or steps other than those stated in a claim. The article "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

Claims

CLAIMS:
1. A system (10) for determining a personal health related risk (25, 312), the system (10) comprising: a sensor input (11) for receiving sensor data related to a health parameter of a patient, a sensor data analysis unit (13) for analyzing and classifying the received sensor data, a context input (12) for receiving contextual data, and a reasoning unit (14) for performing a statistical analysis of the contextual data and of an output of the sensor data analysis unit (13) and for, based on the statistical analysis, determining the personal health related risk (25, 312).
2. A system (10) for determining a personal health related risk (25, 312) as claimed in claim 1, wherein: the personal health related risk (25, 312) is a melanoma risk (25), the sensor data comprises an image of a skin lesion of the patient, the sensor data analysis unit (13) comprises an image analysis unit for analyzing and classifying the image.
3. A system (10) for determining a personal health related risk (25, 312) as claimed in claim 1, wherein: the personal health related risk (25, 312) is a risk of falling (312), the sensor data comprises movement data of the patient the sensor data analysis unit (13) comprises a movement data analysis unit for calculating a balance score based on the movement data.
4. A system (10) as claimed in claim 1, wherein the reasoning unit (14) is arranged for using Bayesian statistics for performing the statistical analysis.
5. A system (10) as claimed in claim 1, wherein the output of the sensor data analysis unit (13) comprises a single classification of the personal health related risk (25, 312) based on the received sensor data.
6. A system (10) as claimed in claim 5, wherein the output of the sensor data analysis unit (13) further comprises an accuracy of the classification.
7. A system (10) as claimed in claim 1, wherein the sensor data analysis unit (13) is arranged for extracting multiple parameters from the received sensor data and wherein the output of the sensor data analysis unit (13) comprises the multiple parameters.
8. A system (10) as claimed in claim 1, further comprising a memory (15) for storing historical data, such as received sensor data, the output of the sensor data analysis unit (13), at least part of the contextual data and/or the personal health related risk (25, 312) and wherein the reasoning unit (14) is arranged for using the historical data for performing the statistical analysis.
9. A system (10) as claimed in claim 1, wherein the contextual data comprises epidemiological data.
10. A system (10) as claimed in claim 9, wherein the epidemiological data comprises a skin type, an age, a gender, a sun-burn history, a hair color, a number of nevi, a personal or family disease history or a part of the body comprising the skin lesion.
11. A method of determining a personal health related risk (25, 312), the method comprising the steps of: receiving sensor data related to a health parameter of a patient, analyzing and classifying the received sensor data, receiving contextual data, performing a statistical analysis of the contextual data and the analyzed and classified sensor data, and based on the statistical analysis, determining the personal health related risk (25, 312).
12. A computer program product, which program is operative to cause a processoro perform a method as claimed in claim 11.
PCT/IB2009/052663 2008-06-27 2009-06-22 System and method for determining a personal health related risk WO2009156936A2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP08159152 2008-06-27
EP08159152.1 2008-06-27

Publications (2)

Publication Number Publication Date
WO2009156936A2 true WO2009156936A2 (en) 2009-12-30
WO2009156936A3 WO2009156936A3 (en) 2010-04-01

Family

ID=41076671

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2009/052663 WO2009156936A2 (en) 2008-06-27 2009-06-22 System and method for determining a personal health related risk

Country Status (1)

Country Link
WO (1) WO2009156936A2 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010099619A1 (en) * 2009-03-05 2010-09-10 Quantum Dental Technologies Inc. Method of assessing oral health risk
WO2011091059A1 (en) * 2010-01-19 2011-07-28 Masimo Corporation Wellness analysis system
JP2013121440A (en) * 2011-12-12 2013-06-20 Jvc Kenwood Corp Health management system
CN106599553A (en) * 2016-11-29 2017-04-26 中国科学院深圳先进技术研究院 Disease early-warning method and device
EP3346402A1 (en) 2017-01-04 2018-07-11 Fraunhofer Portugal Research Apparatus and method for triggering a fall risk alert to a person
US10024660B2 (en) 2012-08-27 2018-07-17 Universite Du Quebec A Chicoutimi Method to determine physical properties of the ground

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002087431A1 (en) * 2001-05-01 2002-11-07 Structural Bioinformatics, Inc. Diagnosing inapparent diseases from common clinical tests using bayesian analysis
US20060269111A1 (en) * 2005-05-27 2006-11-30 Stoecker & Associates, A Subsidiary Of The Dermatology Center, Llc Automatic detection of critical dermoscopy features for malignant melanoma diagnosis
WO2008064120A2 (en) * 2006-11-17 2008-05-29 Amigent, Inc. Method for displaying measurements and temporal changes of skin surface images
US20080140708A1 (en) * 2002-01-31 2008-06-12 Oren Fuerst System and method for providing a computer aided medical diagnostic over a network

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002087431A1 (en) * 2001-05-01 2002-11-07 Structural Bioinformatics, Inc. Diagnosing inapparent diseases from common clinical tests using bayesian analysis
US20080140708A1 (en) * 2002-01-31 2008-06-12 Oren Fuerst System and method for providing a computer aided medical diagnostic over a network
US20060269111A1 (en) * 2005-05-27 2006-11-30 Stoecker & Associates, A Subsidiary Of The Dermatology Center, Llc Automatic detection of critical dermoscopy features for malignant melanoma diagnosis
WO2008064120A2 (en) * 2006-11-17 2008-05-29 Amigent, Inc. Method for displaying measurements and temporal changes of skin surface images

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010099619A1 (en) * 2009-03-05 2010-09-10 Quantum Dental Technologies Inc. Method of assessing oral health risk
US8439681B2 (en) 2009-03-05 2013-05-14 Quantum Dental Technologies Method of assessing oral health risk
WO2011091059A1 (en) * 2010-01-19 2011-07-28 Masimo Corporation Wellness analysis system
GB2490817A (en) * 2010-01-19 2012-11-14 Masimo Corp Wellness analysis system
JP2013121440A (en) * 2011-12-12 2013-06-20 Jvc Kenwood Corp Health management system
US10024660B2 (en) 2012-08-27 2018-07-17 Universite Du Quebec A Chicoutimi Method to determine physical properties of the ground
CN106599553A (en) * 2016-11-29 2017-04-26 中国科学院深圳先进技术研究院 Disease early-warning method and device
EP3346402A1 (en) 2017-01-04 2018-07-11 Fraunhofer Portugal Research Apparatus and method for triggering a fall risk alert to a person
WO2018127506A1 (en) 2017-01-04 2018-07-12 Fraunhofer Portugal Research Apparatus and method for triggering a fall risk alert to a person

Also Published As

Publication number Publication date
WO2009156936A3 (en) 2010-04-01

Similar Documents

Publication Publication Date Title
Das A comparison of multiple classification methods for diagnosis of Parkinson disease
Dawson et al. Comparison of measures to assess outcomes in total hip replacement surgery.
Eftekhar et al. Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data
ES2659945T3 (en) Based monitoring residues of human health
Stone et al. Unobtrusive, continuous, in-home gait measurement using the Microsoft Kinect
Howcroft et al. Review of fall risk assessment in geriatric populations using inertial sensors
Clifton et al. Predictive monitoring of mobile patients by combining clinical observations with data from wearable sensors
Gargeya et al. Automated identification of diabetic retinopathy using deep learning
RU2650586C2 (en) Health monitoring system for calculating a total risk score
KR101632308B1 (en) Method and apparatus for providing blood glucose management information
Staudenmayer et al. An artificial neural network to estimate physical activity energy expenditure and identify physical activity type from an accelerometer
Johnson et al. Machine learning and decision support in critical care
Polat et al. Diagnosis of heart disease using artificial immune recognition system and fuzzy weighted pre-processing
JP2007526016A (en) System and method for automatic local myocardial evaluation of cardiac imaging
Agostini et al. Segmentation and classification of gait cycles
Bode et al. Issues in the development of an item bank
Khandoker et al. Wavelet-based feature extraction for support vector machines for screening balance impairments in the elderly
Oliver et al. Utility of accelerometer thresholds for classifying sitting in office workers
Capela et al. Feature selection for wearable smartphone-based human activity recognition with able bodied, elderly, and stroke patients
Fialho et al. Data mining using clinical physiology at discharge to predict ICU readmissions
Mileo et al. Support for context-aware monitoring in home healthcare
Elgendi et al. Systolic peak detection in acceleration photoplethysmograms measured from emergency responders in tropical conditions
Schwenk et al. Wearable sensor-based in-home assessment of gait, balance, and physical activity for discrimination of frailty status: baseline results of the Arizona frailty cohort study
Zingmond et al. The role of socioeconomic status on hip fracture
Liu et al. Computational methods for estimating energy expenditure in human physical activities

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09769744

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase in:

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09769744

Country of ref document: EP

Kind code of ref document: A2