CN104956391A - Clinical dashboard user interface system and method - Google Patents

Abstract

A dashboard user interface method comprises displaying a navigable list of at least one target disease, displaying a navigable list of patient identifiers associated with a target disease selected in the target disease list, displaying historic and current data associated with a patient in the patient list identified as being associated with the selected target disease, including clinician notes at admission, receiving, storing, and displaying review's comments, and displaying automatically-generated intervention and treatment recommendations.

Description

Clinical instrument panel user interface system and method

field

The disclosure relates to a kind of clinical instrument panel user interface system and method, and relates to the field of disease mark and monitoring particularly.

Background technology

The challenge that hospital is faced now is the primary disease of identified patient as early as possible, can dispose suitable intervention immediately.The some diseases of such as acute myocardial infarction (AMI) and pneumonia and so on requires standard operation immediately in 24 hours of diagnosis or approach.Other diseases so serious but still require multiple health institution conscientiously adhere to medium and long term nursing (care) plan.

Joint committee (the Hospital Accreditation mechanism ratified by medical insurance and Medicaid Service center (CMS)) have developed the core measure of the treatment measures clearly set forth.These measures depend on the standard that can cause punishing the CMS of bad performance.Such as, the measure set by acute myocardial infarction is comprised:

Measure ID is set	The short name of measure
		AMI-1	To aspirin during institute
AMI-2	Prescription aspirin when leaving hospital
		AMI-3	For ACEI or ARB of LVSD
AMI-4	Adult's smoking cessation consulting/teach
		AMI-5	Prescription beta-blocker when leaving hospital
AMI-7	Meta fibrinolysis time (Median Time to Fibrionolysis)
		AMI-7a	The thromboembolism treatment accepted in 30 minutes of arrival hospital
AMI-8	The meta main PCI time (Median Time to Primary PCI)
		AMI-8a	The main PCI accepted in 90 minutes of arrival hospital
AMI-9	Inpatient mortality ratio (12/31/2010 effectively retrieves)
		AMI-10	Prescription Pitavastatin when leaving hospital

Up to now, after patient leaves hospital from health care institution, complete for the movable great majority report of answerable measure (measure) and monitoring.Mark and the delay understanding specific intervention aspect usually make to correct any situation.Administrators of the hospital are also difficult to determine that hospital has every day and meet core measure how well.Clinician needs in the whole while in hospital in real time or close to checking the patient progress and nursing that comprise clinician's notes in real time, and this measure will be reminded towards meeting these core measure and will action (approach and monitoring) in medical control team and doctor.

Case Executive Team is inconvenient for the real-time disease state of tracking patient.The ability having carried out this measure with the clear picture of the notes of clinician (because they arrive with new information in patient's while in hospital and change in real time) applies to focus on by increasing team the ability that (focused) intervene as early as possible, and follow or change these approach (pathway) as required in whole patient's while in hospital, thus improve nursing quality and safely, reduce readmission outside the plan and adverse events and improve patient's result.Present disclosure describes exploitation for identify and to being in the highest to be in hospital again and the patient of other bad clinical events risk carries out the software of risk stratification, and present the instrument panel user interface of data in clear and understandable mode to user.

Accompanying drawing explanation

Fig. 1 is the simplified block diagram of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring system and method;

Fig. 2 is the time shaft figure of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring system and method;

Fig. 3 is the simplification logic diagram of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring system and method;

Fig. 4 is the simplified flow chart of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring system and method;

Fig. 5 is the simplified flow chart/block diagram of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring system and method;

Fig. 6 is the simplified flow chart of the exemplary embodiment according to instrument panel user interface system of the present disclosure and method;

Fig. 7 is with instrument panel user interface system with the simplified flow chart of the mutual exemplary embodiment of the typical user of method according to of the present disclosure;

Fig. 8 is the exemplary screen shots according to instrument panel user interface system of the present disclosure and method;

Fig. 9 comments on the instrument panel user interface system of window and the exemplary screen shots of method according to display of the present disclosure drop-down (drop); And

Figure 10 is according to the display viewing comment instrument panel user interface system of window of the present disclosure and the exemplary screen shots of method.

Embodiment

Fig. 1 is the simplified block diagram of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring system 10.Dlinial prediction and monitoring system 10 comprise computer system 12, and computer system 12 is suitable for receiving the patient of relevant requirements nursing or the various clinical and non-clinical data of individual.Such as, various data comprise from hospital and health care entity 14, non-medical healthcare entity 15, health and fitness information exchange 16 and society to healthy (social-to-health) message exchange and social service agency 17 real-time stream and history or store data.These data are for determining the disease risks mark of selected patient, and they can be accepted as, and their particular case and demand formulate better and customize more the intervention of target, treatment and nursing.System 10 be best suited for mark need intensive be in hospital and/or the particular patient of treat-and-release to avoid the serious adverse effect of specified disease and to reduce admission rate again.It should be noted that computer system 12 can comprise operating into and transmit data and the one or more Local or Remote computer servers communicated via wired with wireless communication link and computer network.

The data received by dlinial prediction and monitoring system 10 can comprise electronic medical record (EMR), and electronic medical record comprises clinical and non-clinical data.EMR clinical data can receive from entity (such as, hospital, clinic, pharmacy, laboratory and health and fitness information exchange), and it comprises: vital sign and other physiological datas; The data be associated with the comprehensive or concentrated history of being undertaken by doctor, nurse or UnitedHealth professional and physical examination; Medical history; Irritated and bad medical treatment reaction before; Family's medical history; History of operation before; Emergency ward record; Drug use administration record; Cultivation results; Dictate clinical notes and record; Gynaecology and obstetrics' history; Mental status examination; Vaccine inoculation recording; Radiophotography checks; The visual operation of invasive; Psychiatric treatment history; Tissue specimen before; Laboratory data; Hereditary information; The notes of doctor; Networked devices and monitor (such as, blood pressure device and blood glucose meter); Information taken in by medicine and replenishers (supplement); Test with concentrated genotype.

EMR non-clinical data can comprise, such as, and society, behavior, life style and economic data; The type of occupation and character; Duty history; Medical insurance information; Hospital's Land use models; Movable information; Cause addiction substance migration; Occupation chemical contact; The frequency of doctor or health system contact; The position of change of residence and frequency; Prediction examination Health questionnaire (such as patient health questionnaire (PHQ)); Personality is tested; Census and consensus data; Surrounding environment; Diet; Sex; Marital status; Educational background; The Distance geometry quantity of household or nursing aide; Address; Housing conditions; Social media data; And level of education.Non-clinical patient data can comprise the data inputted by patient further, such as, inputs or be uploaded to the data of social media web site.

The additional source of EMR data or equipment can provide, such as, such as, laboratory result, medicine distribute and change, EKG result, radiation take down notes, every day weight readings and every day blood glucose test result.These data sources can from the zones of different of hospital, clinic, patient care facility, patient home's monitoring equipment, and other useful clinical or health care source.

As shown in Figure 1, patient data source can comprise non-healthcare entity 15.These are the entity or the tissue that are not considered to Traditional health care supplier.These entities 15 can provide non-clinical data, and this non-clinical data comprises, such as, and sex; Marital status; Educational background; Community and religious organizations participate in; The Distance geometry quantity of household or nursing aide; Address; The socioeconomic data that census tracts (tract) position is reported with the census in this district; Housing conditions; The quantity that house address changes; The frequency that house address changes; To the requirement of government's Living aids; Carry out and keep the ability that medical treatment is preengage; The independence of activities of daily living; Seek medical assistance hour; Seek the position of medical services; Sensory disturbance; Cognitive disorder; Action obstacle; Level of education; Occupation; And for local and national distribution of earnings with the economic situation of absolute and relative item form; Climatic data; Register with health.This type of data source can be further provided with closes the richness information perspicacious of patient lifestyle, such as, the quantity of kinsfolk, relation condition, can help the individual of care of patients and can the Health and Living mode preference of unhealthful result.

Dlinial prediction and monitoring system 10 can exchange (HIE) 16 from health and fitness information further and receive data.HIE is the tissue transferring healthcare information in region, community or hospital system across tissue ground electronics.Clinical and non-clinical patient data is shared between the healthcare entity of the growing one-tenth of HIE in city, country, region or umbrella sanitation system.Data can be derived from a lot of source, such as hospital, clinic, consumer, payer, doctor, laboratory, Outpatient Dispensary, outpatient service center, home for destitute and country or publilc health mechanism.

Healthcare entity is connected to the community organization specifically not providing health care service by the subset of HIE, such as non-government charity, social service agency and municipal agencies.Dlinial prediction and monitoring system 10 can exchange 17 from these community service tissues and society to health and fitness information and receive data, these data can comprise, such as, relevant daily life technical ability, go the traffic capacity of medical reservation, employment aid, training, drug abuse rehabilitation, consulting or removing toxic substances, rent and government utility assistance, homeless state and the acceptance to service, subsequent medical, mental health services, meals and nutrition, food storage (pantry) is served, house is helped, temporary shelter, family health care is accessed, domestic violence, observe reservation, Discharge Guide, Medicine prescriptions, medication guide, neighbouring state, with the information of following the tracks of the ability of recommending and preengage.

Another source of data comprises social media or social networking service 18, such as, FACEBOOK and GOOGLE+web website.The individual that this type of source can provide the quantity of such as kinsfolk, relation condition, mark can help care of patients and can the information of Health and Living mode preference and so on of unhealthful result.Such as, when individual allows, these social media data can be received from web site, and along with the renewal of user's input state, some data can directly from the computing equipment of user.

These non-clinical patient datas provide for patient the comprehensive healthcare environment of entirety more real and describe accurately.Strengthen with this non-clinical patient data, what performed by native system to be in hospital or the analysis and prediction modeling of the recurrent high risk patient of disease becomes more sane and accurately for identifying to be in again.

System 10 is suitable for receiving user preference and system configuration data from the computing equipment (mobile device, flat computer, laptop computer, desk-top computer, server etc.) 19 of clinician in a wired or wireless manner further.Computing equipment is equipped to display system instrument panel and/or another graphic user interface to present system data and report.Such as, clinician (health worker) can generate the Patient list with the highest risk of heart failure mark (such as, front n or front x%) at once.Graphic user interface be suitable for further receive user (health worker) input hobby and configuration.Data can web page, message based on web, text, video information, Multimedia Message, text message, Email and various suitable mode and form form be transmitted, present and be shown to clinician/user.

As shown in Figure 1, dlinial prediction and monitoring system 10 can receive and to flow out in real time or from the data of history, or from the data in batch of each data source.And received data can be stored in data and store in 20 or process data when first not storing data by system 10.Various form can be according to various agreement (comprising CCD, XDS, HL7, SSO, HTTPS, EDI, CSV etc.) in real time with the data stored.Data can be encrypted or protect in other suitable modes.Data obtain (poll) by system 10 from each data source or data can be pushed to system 10 by data source.As an alternative or additional, according to schedule time table or data can be received with batch processing as required.Data store 20 can comprise one or more home server, storer, driver and other suitable memory devices.As an alternative or additional, data can be stored in cloud data center.

Computer system 12 can comprise can be arranged in this locality or the multiple computing equipments in cloud computing field, and computing equipment comprises server.The safety practice of known or later exploitation now or host-host protocol can be adopted to come encrypted or otherwise protect at computer system 12 and the data data routing stored between 20.

Fig. 2 is the time shaft figure of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring system and method.Exemplarily, time shaft figure is for illustrating that how applying clinical prediction and monitoring system and method 10 lead with the readmission reduced about congestive heart failure.Most of United States Hospital makes great efforts self-control (contain) the again admission rate relevant with congestive heart failure.Although much research has found careful discharge planning, nursing supplier coordinates and some combinations of deeply consulting can prevent from being follow-uply in hospital again, be difficult to successfully realize and maintenance at typical U.S. hospital.The all heart failure patients of unified registration, high strength nursing transient program requires the degree of depth of the case management resource that many systems (especially safety net hospital) do not reach.Dlinial prediction and monitoring system and method 10 are suitable for specified disease and situation (30 days readmissions such as in patients with congestive heart failure) accurately risk stratification.

In 24 hours of patient admits, analyze history about the storage of patient with real time data to identify the disease specific relevant with patient and situation, such as congestive heart failure by dlinial prediction and monitoring system and method 10.Further, system 10 calculates the risk score of the congestive heart failure of this particular patient in 24 hours of being admitted to hospital.If the risk score of the congestive heart failure of this particular patient exceedes particular risk threshold value, then in the list of presenting to the high-risk patient of intervening co-ordination team, identify this patient.The process being used for disease mark and risk score calculating is below described in more detail.

Dlinial prediction and monitoring system and method 10 can operate to show, transmit and otherwise present high-risk patient list to intervention co-ordination team, and this intervention co-ordination team comprises other staff involved by doctor, Physician's Assistant, case management person, patient education person, nurse, social worker, kinsfolk and patient care or individual.Present device and can comprise the Email, text message, Multimedia Message, speech message, web page, fax, the sense of hearing and visual alarm etc. that are transmitted by many suitable electronics or portable computing device.Intervene co-ordination team and then priorization can provide target inpatient and treatment to the intervention of most high-risk patient.Dlinial prediction and monitoring system and method 10 can present plan further automatically to comprise intervention and the treatment option of suggestion.The educational counseling in heart failure that some intervention plans can comprise detailed be in hospital clinical assessment and patient's nutrition, medication, case management person and start in early days in the while in hospital patient.Intervene co-ordination team and can carry out orderly clinical and social interaction of being in hospital immediately.In addition, this plan can comprise clinical and social outpatient service intervention and dispose the nursing after leaving hospital and support plan.

High-risk patient is also assigned one group of high strength outpatient service intervention.Once target patient is left hospital, then start outpatient service intervention and nursing.This intervention can comprise from the Effect of follow-up visit by telephone in 48 hours of case management person (such as, nurse) of patient; The appoint reminder of doctor and medication upgrade; The clinic case management of 30 days; Reservation is followed up a case by regular visits in clinic in 7 days of leaving hospital; Follow-up division of cardiology's reservation (if necessary); Access with follow-up primary care.Find that successfully intervention is known minimizing readmission program and policy based on being designed to significantly reduce the 30 days readmissions be associated with congestive heart failure.

Dlinial prediction continues to receive about the clinical and non-clinical data relevant with being designated high risk patient is treated and intervention plan to improve further to diagnose and revise or strengthen with patient in the while in hospital with method 10 with monitoring system after hospital leaves hospital, if necessary.

Patient from after hospital leaves hospital, dlinial prediction and monitoring system and method 10 according to electronic medical record, case management system, community service entity and as above other data sources continue monitoring patient intervention states.Dlinial prediction and monitoring system and method 10 also can directly and ward, case management person and patient alternately to obtain additional information and to impel action.Such as, dlinial prediction and monitoring system and method 10 can notify that the his or her patient of doctor returns hospital, and the message that then doctor can send pre-formatting to system guide this notifications specific cases to manage Executive Team.In another example, dlinial prediction and monitoring system and method 10 can recognize that the reservation that patient misses doctor is not rearranged.This system can send text message to patient and rearrange reservation to remind patient.

Fig. 3 is the simplification logic diagram of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring system and method 10.Because system 10 receives and extracts data with countless forms from many incoherent sources according to different agreement, first the data imported into must stand multi-step process before they are suitably analyzed and use.Dlinial prediction and monitoring system and method 10 comprise data integration logic module 22, and this data integration logic module 22 comprises data extraction process 24, data scrubbing process 26 and data manipulation process 28 further.Although it should be noted that data integration logic module 22 is shown as having different process 24-28, these only complete for illustration of object and can walk abreast, iteration and alternatively perform these process.

Data extraction process 24 uses various technology and agreement directly or by the data source of the Internet in real time or in history autoexec to extract real time clinical and non-clinical data.Preferably in real time, data scrubbing process 26 " cleaning " or preprocessed data, make structural data be in standard format also for the natural language processing (NLP) performed in disease/risk logic module 30 is prepared non-structured text.They are also converted to the form (such as, be calculate object, text data field converts numerical value to) of expectation by data that system also can receive " clean ".

Data manipulation process 28 can relative to metadata dictionary analyze specific data feedback expression and determine particular data feedback whether should be re-equipped put or by alternate data feed back replaced.Such as, given hospital EMR can store the concept of " maximum kreatinin " by different way.Data manipulation process 28 can carry out inferring determining that the best is represented the concept of " kreatinin " that define in metadata dictionary and determines whether feedback reaches maximal value (such as, selection mxm.) by specific for needs reshuffling by which particular data from EMR.

Then data integration logic module 22 makes pretreated data transfer to disease/risk logic module 30.Disease risks logic module 30 can operate and become each patient and calculate the risk score that is associated with identified disease or situation and identify those patients that should accept target intervention and nursing.Disease/risk logic module 30 comprises mark/identification process 32 again, go mark/again identification process 32 be suitable for removing all shielded health and fitness informations according to HIPAA standard before by internet transmissions data.It is also suitable for identification data again.Removable and the shielded health and fitness information added back can comprise, such as, name, telephone number, fax number, e-mail address, social security number, medical record number, health plan beneficiary number, account, certificate or license number, license plate number, device number, URL, comprise street address, city, county, district, all regions that the ratio state (state) of postcode and their equivalent geocoding is little are segmented (except 3 initial numerals of postcode, if disclose available data according to from the current of the Census Bureau), Internet Protocol number, biological attribute data, with other unique identifying number any, feature, or code.

Disease/risk logic module 30 comprises disease identification process 34 further.Disease identification process 34 is suitable for the interested one or more disease or the situation that identify each patient.Disease identification process 34 considers such as laboratory order, laboratory is worth, clinical text and describe notes and other clinical informations and historical information and so on data to determine that patient has the probability of specified disease.In addition, during disease mark, clinical and the non-clinical data of destructuring carries out natural language processing to determine that doctor thinks general disease or various diseases, can during many skies iteratively implementation 34 to become more confident and higher confidence that is that set up in disease mark in diagnosis along with doctor.Patient data that is new or that upgrade can not support the disease of previous identification, and patient will automatically remove from this list of diseases by system.The model of rule-based model and Corpus--based Method study is combined in natural language processing.

Disease identification procedure 34 utilizes the mixture model of natural language processing, and this mixture model is combined with the model of rule-based model and Corpus--based Method study.During natural language processing, original unstructured data (such as, the notes of doctor and report) is first through being called the process of symbolism (tokenization).The elementary cell of symbolism process by using the separator (such as, punctuation mark, space or capitalization are write) of definition text to be divided into information with the form of single character or phrase.Use rule-based model, according to determining that the pre-defined rule of implication identifies and these elementary cells of appreciation information in metadata dictionary.Use Corpus--based Method learning model, disease identification process 34 quantizes relation and the frequency of word and phrase form, and then Using statistics algorithm processes them.Use machine learning, Corpus--based Method learning model produces based on the pattern repeated and relation infers.Disease identification process 34 performs the natural language processing function of multiple complexity, comprises Text Pretreatment, lexical analysis, syntax parsing, semantic analysis, the expression of process multiword, word sense disambiguation and other functions.

Such as, if the notes of doctor comprise following: " 55yo m c h/o dm, cri. has adib rvr now, chfexac, and rle cellulitis carries out 10W, tele ".Then data integration logic 22 can operate journey and these notes is translated as: 55 years old male sex, there is the history of diabetes, chronic renal insufficiency, atrial fibrillation, the congestive heart failure had now with rapid ventricular reaction increases the weight of and right lower extremity cellulitis, will carry out 10 West and carry out continuous cardiac monitoring.

Example before continuation, disease identification process 34 is suitable for determining following aspect further: 1) patient is specific is admitted to hospital due to atrial fibrillation and congestive heart failure; 2) owing to there is rapid ventricular speed, therefore atrial fibrillation is serious; 3) cellulitis is at right lower extremity; 4) continuous cardiac monitoring or remote measurement are carried out to patient; With 5) patient seems to have diabetes and chronic renal insufficiency.

Disease/risk logic module 30 comprises forecast model process 36 further, and forecast model process 36 is suitable for the risk predicting interested specified disease or situation according to one or more forecast model.Such as, if the grade of the risk of the following readmission of all patients be admitted to hospital due to heart failure is at present expected to determine by hospital, then forecast model in heart failure can be selected to process patient data.But, if the risk class determining all medical patient caused due to any reason is expected by hospital, then all reason readmissions forecast model can be used to process patient data.As another example, if hospital expects that mark is in those patients of the risk of short-term and long term diabetes complications, then diabetes forecast model can be used to come for these patients.Other forecast models can comprise HIV readmission, diabetes identify, the risk of cardiopulmonary all standing, kidney trouble progress, acute coronary syndrome, pneumonia, cirrhosis, all reason disease have nothing to do readmission, colon cancer path compliance etc.

Example before continuing to use, forecast model for congestive heart failure can consider one group of risk factors or variable, comprise the worst-case value for laboratory and vital sign variable, such as: the partial pressure of albumin, total bilirubin, creatine kinase, kreatinin, sodium, blood urea nitrogen, carbon dioxide, white blood cell count, Troponin I, glucose, international normalized ratio, brain natriuretic peptide, pH, temperature, pulse, diastolic pressure and systolic pressure.And, also consider non-clinical factor, such as, the quantity that in upper one year, home address changes, risky healthy behavior (such as, use forbidden drug or material), the history of the quantity of emergency department visits, depression or anxiety and other factors in upper one year.How forecast model appointment classifies and weighting to each variable or risk factors, and calculates the prediction probability of readmission or risk score.By this way, dlinial prediction and monitoring system and method 10 can carry out layering (stratify) to the risk of each patient arriving hospital or another health institution in real time.Therefore, automatically mark is in those patients the most high risk, makes to start (institute) target intervention and nursing.The risk score of all patients of specified disease or situation is comprised from an output of disease/risk logic module 30.In addition, module 30 can arrange patient according to risk score, and is provided in the mark of those patients at list top place.Such as, hospital can expect that mark has the most high risk front 20 patients for congestive heart failure readmission, and has heart cardiopulmonary all standing to have the most high risk patient of front 5% most in following 24 hours.Can the other diseases of usage forecastings model identification and situation comprise, such as, HIV readmission, diabetes mark, kidney trouble progress, intestinal cancer closed set (continuum) screening, meningitis management, soda acid management, anti-freezing management etc.

Disease/risk logic module 30 can comprise spatial term module 38 further.Spatial term module 38 is suitable for receiving the output (such as, the risk score of patient and risk variable) from prediction module 36, and " translation " data are to present the high risk evidence that patient is in this disease or situation.Therefore this module 30 has been identified as the high risk additional information of specified disease or situation to intervening the co-ordination team why patient that provides support.By this way, intervene co-ordination team can set objectives better be in hospital and outpatient service intervention and treatment plan to solve the concrete condition of patient.

Disease/risk logic module 30 comprises artificial intelligence (AI) model adjustment process 40 further.Artificial intelligence model adjustment process 38 uses machine learning techniques to use adaptive learning capacity.The ability that oneself reshuffles can make system and method 10 enough flexibly and strong adaptability to detect and the trend combined in lower floor's (underlying) patient data of the precision of prediction that can affect given algorithm or colony or difference.Artificial intelligence model adjustment process 40 with the precision result be improved, thus can allow to select most precise statistical method, variable counting, variables choice, alternately item, weight and the interception (intercept) for local health system or clinic by the forecast model selected of retraining periodically.Artificial intelligence model adjustment process 40 three kinds of exemplary approach automatically can be revised or improves forecast model.The first, it can regulate clinical and prediction weight that is non-clinical variable when not having manpower to supervise.The second, it can regulate the threshold value of concrete variable when not having manpower to supervise.3rd, artificial intelligence model adjustment process 40 can be assessed when not having manpower supervise and is present in data feedback but is not used in the new variables of forecast model, and this measure can cause the precision of improvement.Artificial intelligence model adjustment process 40 can by the result actually observed of event compared with the result of prediction, then individually in analytical model to the contributive variable of incorrect result.Then it can weigh again to the contributive variable of incorrect result, makes in following iteration, and these variablees are less may have contribution to error prediction.By this way, artificial intelligence model adjustment process 40 is suitable for the concrete clinical setting of applying based on it or colony reshuffles or regulates forecast model.And non-manual to reshuffle or revise forecast model be necessary.Artificial intelligence model adjustment process 40 also can have for forecast model being adjusted to different health system, colony and geographic area in proportion in fast time regimes.

As the example how artificial intelligence model adjustment process 40 works, can periodically assess sodium variation coefficient, to determine or to identify that the relative weighting about the abnormal sodium laboratory result of new colony should from 0.1 to 0.12 change.Along with time lapse, artificial intelligence model adjustment process 38 checks whether the threshold value that should upgrade and receive.Can determine, in order to make the abnormal threshold level receiving laboratory result be predictability for readmission, this exception is received laboratory result and should be changed from such as 140 to 136mg/dL.Finally, artificial intelligence model adjustment process 40 is suitable for checking whether and should upgrades forecast set (variable and the mutual list of variable) to reflect the change of PATIENT POPULATION and clinical practice.Such as, variable of receiving can be replaced by NT-por-BNP protein variable, and this NT-por-BNP protein variable does not have prior predicted model to consider.

To be presented by data from the result of disease/risk logic module 30 and to be provided to hospital personnel (such as, intervening co-ordination team) and other keepers with system configuration logic module 42.Data present logic module 42 and comprise instrument panel interface 44, and instrument panel interface 44 is suitable for providing the information of the performance about dlinial prediction and monitoring system and method 10.User (such as, hospital personnel, supvr and intervention co-ordination team) can find them to be carried out the particular data found by simply and clearly vision guided navigation prompting, icon, window and equipment.Such as, interface can further in response to listening order.The quantity of the patient be admitted to hospital every day due to hospital can be inundant (overwhelming), therefore maximum efficiency the simple graph interface of reducing user's navigation time expects.Visual cues (such as, readmission, leaves ICU (out-of-ICU), heart arrest, diabetic complication, and other) is preferably presented when evaluation problem.

The varied opinions that panel board user interface 44 allows to calculate the data of the extraction from intrasystem operating database and risk score, report and the interaction request presented, comprise, such as, and the synoptic diagram of the Patient list in concrete nursing position; The detailed explanation of the component in each subfraction; The time dependent figure of data of patient or colony is represented; The incidence of the predicted events in special time period and comparing of prediction rate; About the summary texts newspaper cutting of particular patient, laboratory trend and risk score, for helping to give an oral account and prepare history and health check-up report, daily notes, the continuity of signature of nursing notes, notes of performing the operation, discharge abstract, nursing documents to the continuity of outpatient service doctor; Order generates, for automatically generate by this locality nurse supplier's healthcare environment and state and national guidance mandate to be back to doctor's office, outside healthcare provider's network or for being back to hospital or putting into practice the order (order) of electronic medical record; There is provided in data aggregate to conventional medical formula to help to nurse, these data include but not limited to: the creatinine clearance of soda acid calculatings, MELD mark, Child-Pugh-Turcot mark, TIMI risk score, CHADS mark, estimation, body surface area, body mass index, adjuvant, neoadjuvant and metastatic cancer are survived alignment diagram, MEWS mark, APACHE mark, SWIFT mark, NIH Stroke Scale, PORT mark, AJCC pathological staging; Announce the element of the scanning of relational tables or the data of electronic version to create automation data form.

Data present and comprise message transmission (messaging) interface 46 further with system configuration logic module 40, Message passing interface is suitable for such as, and the Form generation output message of the document of the structure of the action of the speech of the transmission of HL7 message, text message transmission, email message transmission, Multimedia Message transmission, web page, web door, REST, XML, Practical computer teaching, the figure comprising risk assessment, numeral and text snippet, prompting and suggestion transmits code.The intervention being generated by system and method 10 or advised can comprise: the risk score report being sent to readmission's risk of the patient for giving prominence to them of the doctor in charge; Be input to the mark report in EMR via new data field, for hospital, cover the crowd regulation use that entity, responsible Nursing talent or health care provide the whole colony in the tissue of other grades in network; The comparison of the integrated risk of the readmission between single hospital or hospital is compared with the risk criteria allowing readmission of hospital and lead; Divide the HL7 message of the automatic combination counting to discharge abstract template, the continuity of nursing document (in the supplier in arranging in hospital or to outside consultant doctor and primary care physicians), convenient readmission's risk transition (transition) that communicates to non-hospital doctor; And the communications subcomponent of set social environment mark, clinical scores and overall risk mark.These marks are by outstanding potential strategy to reduce readmission, and this strategy comprises: generate the list of medications optimized; Allow those medicines on the mark formulary of pharmacy to reduce (out of pocket) expense of sustaining economic losses and to improve the outpatient service meeting drug rehabilitation program; Mark (flagging) nutrition education demand; Mark transportation demand; Assessment house instability is with the needs of mark to home for destitute arrangement, transition house, the 8th part HHS house assistance; Identify bad self-supervision behavior to carry out additional follow-up phone call; Mark causes the bad social networks mark to the suggestion that additional family RN assesses; Mark high substance abuse mark for the consultation of doctors of teaching the rehabilitation of the patient with high abuse problem.

This output can by wirelessly or via LAN, WAN, internet transmissions and be passed to health institution electronic medical record store, consumer electronic devices (such as, pager, text message convey program, mobile phone, panel computer, mobile computer, laptop computer, desk-top computer and server), health and fitness information exchanges and other data storage, database, equipment and user.System and method 10 automatically can generate, transmit and present such as there is risk score high-risk patient list, the text of spatial term, report, the action of suggestion, alarm, the continuity of nursing documents, mark, appoint reminder and survey and so on information.

Data present and comprise system configuration interface 48 further with system configuration logic module 40.Local clinical preference, knowledge and method can be directly provided as the input of forecast model by system configuration interface 46.This system configuration interface 46 allows mechanism or health system directly to arrange or reset variable threshold, other parameters predicted in weight and forecast model.System configuration interface 48 preferably includes the graphic user interface being designed to minimum user navigation time.

Fig. 4 is the simplified flow chart of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring method 50.Method 50 receives from each provenance and with the structuring relevant with particular patient of multiple different-format with destructuring is clinical and non-clinical data, as depicted in element 52.Data security methods that is known or later exploitation now can be used to encrypt or protect these data.In frame 54, the data that method 50 pre-service receives, such as, data extraction, data scrubbing and data manipulation.Other data processing techniques of known and later exploitation now can be used.In frame 56, data processing method (such as, natural language processing) and other suitable technology can be used for the meaning translating or otherwise understand data.In block 58, by analyzing pretreated data, identify the interested one or more disease relevant with each patient or situation.In block 60, method 50 is applied one or more forecast model and is analyzed data further and the one or more risk score calculating each patient relevant with the disease of mark or situation.In frame 62 and 64, display is had one or more lists of the disease of each mark or those patients the most high risk of situation are generated, transmit and be otherwise presented to hospital personnel, such as, intervene the member of co-ordination team.Can every day or according to another expect timetable generate these lists.Intervening co-ordination team can then prescription (prescribe) following for being in hospital and the target intervention of Clinic Nursing and treatment plan.In frame 66, be designated these patients high risk stand be in hospital and Clinic Nursing while, continuously monitor be designated these patients high risk.Ending method 50 in frame 68.

Clearly identification process is gone in display in the diagram, and in this process, data become uncorrelated with the identity of patient and specify to meet HIPAA.Whenever data by can wired or wireless network link transmission that is damaged and that otherwise required by HIPAA time, data can with the identity decoupling zero of patient.Method 50 is suitable for patient data is combined with the identity of patient again further.

Fig. 5 is the simplified flow chart/block diagram of the exemplary embodiment according to dlinial prediction of the present disclosure and monitoring method 70.Various data are received from multiple incoherent data source 72 place relevant with the particular patient of being in hospital in hospital or health institution.Can the data imported into of real-time reception or the data historical data that can be stored as batch or receive as required.The data imported into are stored in data and store in 74.In block 76, as mentioned above, the data received stand data integration process (data extraction, data scrubbing, data manipulation).Then the pretreated data of gained stand disease logical process 78, in the meantime, perform and go mark, disease mark and prediction modeling.In frame 80 by the risk score of interested disease that calculates for each patient compared with disease risks threshold value.The risk threshold value of each disease and its oneself is associated.If risk score is less than risk threshold value, then process is back to data integration and repeats this process when the new data that new data are associated with patient become available.If risk score is more than or equal to risk threshold value, then in frame 82, the patient with excessive risk mark identified is included in Patient list.In frame 84, then Patient list and other information be associated can be presented in one or more possible modes and intervene co-ordination team, such as transfer to desk-top or mobile device in the mode of text message, Email, web page etc. and show thereon.As indicated in block 88, by this way, notify and activate and intervene the patient of co-ordination team to the identified in list of patient and assess and be in hospital and treat-and-release and nursing.Then this process can provide feedback data to data source 72 and/or be back to data integration 76, and data integration 76 to be in hospital and outpatient service intervention and treatments period continue to monitor patient in his/her target.According to the algorithm of preassignment, continuous monitoring with to be in hospital and during Clinic Nursing the data relevant with patient that generate are (such as, the medicine of institute's prescription and further laboratory result, radiology image etc.), the nursing care plan of the algorithm definition patient of this preassignment.

Fig. 6 is the simplified flow chart of the exemplary embodiment according to instrument panel user interface system of the present disclosure and method 90.In block 92, the data of assess patient as described above, and identify those patients be associated with target disease and supervision situation.Target disease is that patient is in readmission's those diseases to the risk of health institution.The situation monitored refers to those status of patient being shown in health institution and adverse events occurs, such as, and damage and injury.As shown in picture frame 94, by with predetermined probability threshold value relatively come to verify further that relevant specified disease or supervision situation comprise this patient.If meet probability threshold value, then patient is classified or is designated and belongs to list of diseases or condition list.Also refresh display, makes, when user selects specific list of diseases to show, to show this patient in lists, as block 96.This can see in exemplary screen in fig. 8.In this exemplary screen, list in effective congestive heart failure identified in list the list being in the patient of 30 days readmission's risks due to congestive heart failure (CHF).The detailed description of exemplary screen is below provided.

User is printable, transmission and otherwise use shown information, and generates standard or customization report.This report can be main text in essence, or comprises graphical information.Such as, the comparison that the readmission that the readmission that Graphics Report can be plotted in concentrated intervention program the expection for any disease type, situation or classification of registering (enrolled) or unregistered patient leads and observes leads, and the patient of registration is relative to the chart led for the readmission over a period of any disease type, situation or classification of patient abandoning (dropped).The upper total patient of patient, the at the appointed time cycle with probability in heart failure being greater than 95% had relative to do not have in the patient of registration and readmission's window of leaving hospital for 30 days the quantity of the patient of readmission and (by) from leave hospital through number of days.The report of additional exemplary standard-track can identify the patient of all registrations further, for these patients: arrangement is preengage after leaving hospital, arrangement leave hospital after telephone counseling, patient has participated in and has followed up a case by regular visits to reservation, telephone counseling after patient has received and left hospital, patient has received and has filled in medicine prescription, and patient has received traffic certificate (transportation voucher).And, sample report can comprise for registration with unregistered patient for any disease type, event or classification expection lead with the readmission observed compare, the patient of registration for any disease type, event or classification is over a period relative to the patient readmission rate abandoned, there is the patient with probability in heart failure being greater than 95%: the patient that the total patient at the appointed time on the cycle registers relatively, and the patient populations of window Nei Wei readmission of readmission of leaving hospital for 30 days and (by) from leave hospital through number of days.If do not meet probability threshold value in block 94, then the data of patient are reassessed in a timely manner as data that are new or that upgrade become available.

The report of available another type is result optimizing report.These design to help effect of user (supvr) appraisal procedure, set up benchmark and mark to the needs of the change of system and colony's grade to improve the report of nursing result.Report can comprise the data of the effect helping assessment mark high-risk patient.Some data can show the effort of cost, the patient of registration, and how long those patients' reality are subject to the torment of identified disease once.Report can comprise and helps assessment to intervene whether to give the data of correct patient etc. in the correct time.

Along with new, that upgrade or additional patient data become available, as shown in block 98, assess to identify or verify disease/situation to these data.Such as, if data indicate patient to be classified differently now, then patient can be reclassified.Patient also can be identified as additional disease and be included in another list.Such as, at first that is in hospital in 24 hours, patient Jane Doe is designated by system has CHF.Once receive more information (such as, laboratory result and new doctor notes), Jane Doe is designated by system also has AMI.Then Jane Doe will be placed in AMI list, and new diagnosis is identified as AMI patient once available.In addition, Jane Doe will remain in CHF list, but she will be identified as AMI patient in the list.

If there is no new patient data, then, as shown in frame 99, do not make any change to patient class, and the current state of display reflection patient class.Therefore, along with in real time or become available close to real-time patient data, the disease of patient and status classification are reappraised as required and are upgraded.

Fig. 7 is according to the simplified flow chart with the exemplary embodiment of typical user's reciprocal process 100 of instrument panel user interface system and method for the present disclosure.Allow all users of access system must have login security information, the username and password of such as placing on record.As shown at block 102, all visiting demands of system are logged in system by providing correct log-on message.As shown in frame 103, user can select such as, and the parameter of disease type, event, risk class and suitable lattice and so on is selected into generate report for the nursing intervention program of high strength.User can make this selection in any moment after successful login.Such as, user can select particular patient and consult the information relevant with this patient.As indicated at block 104, then user can consult and assess shown information, and this information comprises the physical notes of selected parts (clipped).User some forms can also print, transmit or otherwise use shown information.

Target prediction readmission disease can comprise: congestive heart failure, pneumonia, acute myocardial infarction, diabetes, cardiopulmonary all standing and mortality ratio, cirrhosis readmission, HIV readmission, septicemia and all reasons.Target disease mark can comprise: the chronic renal disease in chronic renal disease, septicemia, supervision, outpatient service, the diabetes in outpatient service and septicemia.What cause due to possible adverse events can comprise for the goal condition of supervising: septicemia, postoperative pulmonary embolism (PE) or Deep vain thrombosis (DVT), postoperative septicemia, postoperative shock, recurrence operation outside the plan, respiratory failure, hypertension, accidental injury, assessment diagnosis, or the communication deficiency in monitoring, omit or mistake, fall, the infection of infection from hospital, wrong medicine patient, Patient identification's problem, leave ICU cardiopulmonary all standing and mortality ratio, chronic renal disease, shock, the initiation of naloxone, the initiation (excess sedation) of anesthesia, hypoglycemic initiation, and unexpected death.

Such as, this assessment can comprise the comment of input about patient.As a part for evaluation process, user can confirm, denies or express the uncertainty about the disease of patient or situation mark or the suitable lattice of intervention program registration.Such as, as indicated at block 106, user can consult the notes and suggestion that are associated with particular patient and confirm to comprise this patient in congestive heart failure list.Such as, user consults the notes of the clinician of the selected parts pointing out key word and phrase, thus helps him or she to find the key message about disease mark by system.Such as " breathe hard ", disease mark that the Key Term of " BNP rising " and " furosemide (Lasix) " and so on helps the CHF of this patient of user rs authentication.If confirm the classification of patient, risk class and suitable lattice grade, then as shown in frame 107, the classification of patient and shown data (except instruction has confirmed except this classification) aspect do not change.User can provide the comment be associated with this confirmation.User comment is stored and by other users in real time or close to checking in real time, thus can allows clear between Team Member and link up timely.User can continue to select report or display parameter in frame 103, or consults and assess patient in frame 104.

Alternatively, user can not agree to include this patient in congestive heart failure list, or expresses uncertain.User can input the comment that the assessment explaining that he or she identify the disease of patient and the disease not agreeing to patient identify.User comment is stored and by other users in real time or close to checking in real time, thus can allows clear between Team Member and link up timely.

If user denies this classification, then as illustrated in block 108, patient is removed from effective list of target disease or situation, and this patient is placed in abandons list (drop list).Deny this classification in response to user, system can additionally show or mark about to information particular list comprising the contributive patient of this patient.Such as, if user denies the disease ID that John Smith is in heart failure, then system can show inquiry further: " Mr. Smith raises because following factor may have CHF:BNP, is short of breath, is admitted to hospital due to decompensation CHF6/9.You determine that you need this patient to remove from effective CHF list? " user need be with no come in response to this inquiry.System can additionally ask ultimate principle (rationale) to intend patient to remove patient from effective list from user.Customer-furnished ultimate principle can be stored and be shown as consultant's comment.User also can indicate uncertainty, and as shown in frame 109, patient is removed from effective list and is placed in watch list for further assessment.Then user can consult and assess the additional patients in same target list of diseases or consult the patient be included in other diseases and condition list.At any point, user some forms can print, transmit and otherwise use shown information, such as generates standard or customization report.

Exemplarily, patient Kit Yong Chen was once identified as CHF patient when being admitted to hospital.After the more data (that is, new laboratory result and new doctor notes) receiving her while in hospital, system by this Patient identification for having AMI.

When being admitted to hospital, clinician takes down notes statement: 52yo women w pmh has CAD, also has the SOB that goes from bad to worse and oedema one month with HTN.1. have difficulty in breathing: possible CHF reduces with the BNP afterload of the rising adopting aCEi and adopts the furosemide of frusemide.O2 statistics stablizes 2) troponin rising: the EKG with contingency model follows the card that serial enzyme carrys out the ROMI cathepsin possible with consulting.After this, clinician takes down notes statement: 52yo women has the CAD of pmh, also has the SOB that goes from bad to worse an and oedema month c CAD with HTN and has LHC and have support prox LAD.1.Troponin rising-NSTEMI, although pt denies that CP – pt has the hx of known CAD, slight troponin leaks the 0.13->0.15->0.09-GreatT.Grea T.GT0.1-pt that is admitted to hospital and gives 325, Plavix load is 300 milligram 1, increase 50mg q6 with heparin gtt-Metop, Coreg – LHC may be changed afterwards and adopt PCI according to division of cardiology today.EP is also discussed and settles 2 for possible ICD.In heart failure, acute and chronic (acute on chronic)-serious diastolic dysfunction starts furosemide 40tid at first because proBNP when HTN drug withdrawal +/-CAD-is admitted to hospital raises 3183-, oedema is greatly improved, and present furosemide 40po bid-TTE completes display: 4 Room expansions, RVH, nml LV thickness, major depression LVEF, LVEF 30%, medium MR, slight TR, AR and PR; Serious diastolic dysfunction, RVSP 52-continue furosemide, lisinopril, Metop – adopt EP and preliminary medical to manage to inquire into AICD and evaluate.

Consultant can assess to take down notes with second of the PIECES disease ID with AMI compared with there is being admitted to hospital of CHF disease ID take down notes, to make great efforts to verify new real-time disease mark.The notes instruction CHF that is admitted to hospital is principal disease.The outstanding term of the key of CHF is indicated to comprise " pmh of CAD " (medical history of coronary artery disease, " SOB " (short of breath), " oedema ", " BNP of rising " to user.Second notes are to user's instruction, although patient has CHF, CAD is the main cause of CHF.Crucial outstanding term (such as, " troponin of rising " and " NSTEMI (non-ST section cardiac asthenia infraction: heart attack) ") gives custom system for AMI being designated the snap view of the Key Term of principal disease.These outstanding Key Terms give user for the real-time or close instrument verifying the change of the system of disease mark in real time.Then user adopts new disease mark to confirm, deny or express uncertainty.In this example, the change identified with outstanding term assessment notes and by accepting disease is verified this change by consultant.Because the main intervention of patient will be used for AMI, therefore patient, the disease identified and risk class will appear in AMI list.

Instrument panel user interface also can indicate the change of risk class.Such as, once return laboratory result (slight raise creatinine and toxicity analyzer (tox screen) cocaine is positive) and affect risk other social factors (due to homeless and not in accordance with receiving restriction) and medical path language queue, this Patient identification can be excessive risk by system.User can follow these in real time and change and the change verifying risk level.

Fig. 8 is the exemplary screen shots 120 according to instrument panel user interface system of the present disclosure and method.Exemplary screen 120 display is identified as because congestive heart failure has multiple patients of the risk of health institution of readmission.The effective congestive heart failure list of exemplary screen displays.The left-hand side of screen is that user can select target disease for consulting and assessing and supervision situation.Target disease is those diseases that also patient can be placed in the risk of health institution of readmission for its assess patient.Registration and supervision situation are those situations of the result that can be the adverse events occurred in health institution.Owing to can be used for the space showing exemplary screen, only show and select several disease and situation, and should be understood that, the patient data for any amount of target disease and situation can be assessed and analyze to this system and method.Instrument panel user interface system can operate into the patient that tissue and display belong to multiple list: effectively list (disease identified or situation), watch list (uncertain) and abandon list (by what deny).User can click any label to check and print any list.Show the multiple data items be associated with each patient in list, such as be admitted to hospital or to institute's date, patient's name, the target disease identified or situation, state (in the concentrated registration intervened in programming), whether confirm that the risk of identified disease and readmission (is expressed as, such as, high, in and low).The type of the data shown for each Patient list can change.It should be noted that the data of the other types be associated with each patient in list can be shown as the bed label of such as each patient and medical record number (MRN) to transmit and otherwise for identified patient.Exemplary screen 120 also can in the name of the top of screen display user and position (doctor, case management person, RN, operation nurse etc.).Be in had adopted today, this week and last week low, in and excessive risk statistical magnitude select heart failure risk patient quantity and be made into table and show in exemplary screen further.

User can click the particular patient shown in lists, and obtains the additional detailed information about this patient.Such as, near the bottom of screen, show clinician's notes (assessment of patient and plan) of selected parts, and key word and phrase are highlighted or otherwise emphasize to indicate and comprise this patient, situation, risk class and suitable lattice grade those texts especially contributive to identified target disease list.The clinician that user can roll through all selected parts be associated with patient takes down notes, and these clinicians notes are chronologically organized, and makes user can consult the progress of disease, diagnosis, assessment and approach.To see in real time due to user or close to checking that these are taken down notes in real time, he or she can verify the assessment of the system of non-structured text clinically.The comment that the confirmation identified with disease or situation that display provides consultant further or deny is associated.

Clearly do not show supplementary features in fig. 8, such as, input one or more search condition to find out the search column of one or more particular patient in order to user can be made.Such as, user can input medical record number, name, admission date, disease type, risk class, event, registration program etc. meet one group of patient of search condition with mark.Search condition based on the standard of other types, such as, can miss those patients preengage after it is left hospital.

Fig. 9 and 10 illustrates abandon comment window and observe the comment instrument panel user interface system of window and the exemplary screen shots 122 and 124 of method.Indicate specific patient not to be taken in list of diseases if user clicks " No ", the reason abandoning window ejects supports the clinical of the decision-making of the classification of the patient denied in AMI list and non-clinical condition to enable user select.User can input the reason do not shown further.Similarly, as example as shown in Figure 10, user can input the probabilistic reason expressed about pneumonia list comprising particular patient.

Display optionally comprises the suggestion and prompting that are generated by system further.These suggestions and prompting can advise evidential intervention options, and this evidential intervention options provides maximum health benefits by patient.The intervention proposed can consider clinical and non-clinical patient variables.In addition, former patient's enrollment results is included (factor) in the intervention of suggestion.The order of PHC automatically can be sent when patient registers in a program, such as, nutrition, medication etc.

System operable described herein becomes in real-time or close utilization (harness) in real time, simplifies, selects and present patient information, prediction and the most high risk patient of mark, mark adverse events, coordination and warning practitioner and monitor patient's result over time and space.Native system improves health care efficiency, helps Resourse Distribute and present the important information causing better patient's result.

The feature in claims is below adopted to set forth the feature being considered to have novelty of the present invention.But, apparent to one skilled in the art to the amendment of exemplary embodiment described above, change and change, and system and method described herein comprises these amendments, changes and change and be not limited to specific embodiment described herein.

Claims (22)

1. an instrument panel user interface method comprises:

What show at least one target disease can navigating lists;

What show the Patient identifier be associated with the target disease selected in described target disease list can navigating lists;

The history that display is associated with the patient be identified as in the Patient list that joins with selected target disease association and current data, the clinician comprised when being admitted to hospital takes down notes;

Receive, store and show the comment of consultant; And

The intervention that display generates automatically and treatment suggestion.

2. instrument panel user interface method as claimed in claim 1, is characterized in that, display history and current data comprise further in real time or take down notes close to showing clinician in real time.

3. instrument panel user interface method as claimed in claim 1, is characterized in that, display history and current data comprise the clinician's notes being presented at the while in hospital further.

4. instrument panel user interface method as claimed in claim 1, is characterized in that, display Patient identifier comprises display patient's name and medical record number.

5. instrument panel user interface method as claimed in claim 1, it is characterized in that, comprise further show the information that is associated with patient can navigating lists, described information comprises patient's name, admission date, the target disease identified, risk class and whether confirms that disease identifies.

6. instrument panel user interface method as claimed in claim 1, is characterized in that, comprises further showing request to consultant and confirming or deny the inquiry that disease identifies.

7. instrument panel user interface method as claimed in claim 1, is characterized in that, display clinician notes comprise display and have clinician's notes disease being identified to the contributive key word emphasized.

8. instrument panel user interface method as claimed in claim 1, it is characterized in that, what show the patient be associated with at least one target disease can comprise the patient showing and have patient data by navigating lists, wherein analyze described patient data by risk logic module, described risk logic module can operate into by described patient data applied forcasting model to determine at least one risk score of being associated with at least one in target disease and to identify at least one high-risk patient of at least one in target disease, described forecast model comprises considers clinical at least one high-risk patient be associated with at least one in target disease with mark with multiple weighted risk variable and the risk threshold value of non-clinical data.

9. instrument panel user interface method as claimed in claim 1, it is characterized in that, the form comprising being selected from least one in the group comprising following item further generates and transmission information: report, graph data, text message, Multimedia Message, instant message, speech message, email message, web page, the message based on web, multiple web page, based on the message of web and text document.

10. instrument panel user interface method as claimed in claim 1, is characterized in that, comprises further generating and at least one mobile device transmission notice and information.

11. instrument panel user interface methods as claimed in claim 1, it is characterized in that, comprise further show at least one situation of being associated with adverse events can navigating lists, and the Patient identifier that is associated with at least one situation of display can navigating lists.

12. instrument panel user interface methods as claimed in claim 1, is characterized in that, comprise further show be associated with target disease effective list, abandon list, watch list and leave hospital in list at least one.

13. 1 kinds of instrument panel user interface methods, comprising:

What show at least one target disease can navigating lists;

What show at least one situation be associated with adverse events can navigating lists;

What show the Patient identifier be associated with the target disease selected in target disease list can navigating lists;

Show and be designated the data that in the Patient list that joins with selected target disease association, patient is associated, comprising the clinical note associating contributive outstanding text had with selected target disease; And

Receive, store and show the comment of consultant, comprise for confirming, denying or express about the probabilistic ultimate principle associated with selected target disease.

14. instrument panel user interface methods as claimed in claim 13, is characterized in that, comprise further in real time or take down notes close to showing clinician in real time.

15. instrument panel user interface methods as claimed in claim 13, is characterized in that, comprise the history and Present clinical physical notes that are presented at the while in hospital further.

16. instrument panel user interface methods as claimed in claim 13, is characterized in that, display Patient identifier comprises display patient's name.

17. instrument panel user interface methods as claimed in claim 13, it is characterized in that, comprise further show the information that is associated with patient can navigating lists, described information comprises patient's name, admission date, the target disease identified, risk class and whether confirms that disease identifies.

18. instrument panel user interface methods as claimed in claim 13, is characterized in that, comprise further and show to consultant the inquiry asking to confirm or deny classification of diseases.

19. instrument panel user interface methods as claimed in claim 13, is characterized in that, display clinician notes comprise showing to have takes down notes the clinician of the contributive key word emphasized of classification of diseases.

20. instrument panel user interface methods as claimed in claim 13, it is characterized in that, what show the patient be associated with at least one target disease can comprise the patient showing and have patient data by navigating lists, wherein analyze patient data by risk logic module, described risk logic module can operate into and apply forecast model to determine at least one risk score of being associated with at least one in target disease and to identify at least one high-risk patient of at least one in target disease to described patient data, described forecast model comprises considers clinical at least one high-risk patient be associated with at least one in target disease with mark with multiple weighted risk variable and the risk threshold value of non-clinical data.

21. instrument panel user interface methods as claimed in claim 13, it is characterized in that, the form comprising being selected from least one in the group comprising following item further generates and transmission information: report, graph data, text message, Multimedia Message, instant message, speech message, email message, the web-page, the message based on web, multiple web page, based on the message of web and text document.

22. instrument panel user interface methods as claimed in claim 13, is characterized in that, comprise further generate and at least one mobile device transmission notice and information.