Diabetes care related process modelling using Process Mining techniques. Lessons learned in the application of Interactive Pattern Recognition: coping with the Spaghetti Effect
Download as PPTX, PDF1 like938 views
The document discusses the use of process mining techniques to model diabetes care processes and improve patient management. It highlights the challenges associated with the 'spaghetti effect' in clinical pathways and advocates for standardized and automated processes. Additionally, the text emphasizes the potential of interactive pattern recognition technologies to enhance the design and deployment of personalized care protocols for diabetes patients.
Diabetes care related process modelling using Process Mining techniques. Lessons learned in the application of Interactive Pattern Recognition: coping with the Spaghetti Effect
- 1. Innovaciones Tecnologías para la Salud y el Bienestar MIlano, 26th August 2015 Diabetes care related process modelling using Process Mining techniques. Lessons learned in the application of Interactive Pattern Recognition: coping with the Spaghetti Effect Carlos Fernandez-Llatas, Antonio Martinez-Millana, Alvaro Martinez-Romero, Jose Miguel Benedí and Vicente Traver
- 2. Where is the pain? - The worldwide prevalence of Diabetes was estimated around 8.3% among adults between 20-80* - Self management improves quality of life of patients - Personalisation requires evidence (EBM) - Evidence needs to be extracted from available data to suggest/define new care processes/Life Assistance Protocols for diabetes management * L. Guariguata, D. R. Whiting, I. Hambleton, J. Beagley, U. Linnenkamp, and J. E. Shaw. Global estimates of diabetes prevalence for 2013 and projections for 2035. Diabetes Research and Clinical Practice, 103(2):137–149, February 2014
- 3. Need to design new care processes /Life Assistance Protocols Process standardization Automation and traceability Measure and optimization Workflows as automation language Complete, unambiguous formal processes Expressivity vs understandability Difficulties in the process design using workflows Complete and explicit definition needed Differences between current and perceived process High time consuming Deep knowledge of representation language
- 5. Process Mining Process Mining Obtain a model from the execution logs Corpus Set of workflow execution logs used to train the model Our Vision: Activity-Based Workflow Mining Parallel Activity-based Log Inference Algorithm (PALIA) … 27/05/2013 19:31:18.65 => i:9 Fin Accion: FirstTriageResult Res: ACUTE 27/05/2013 19:31:18.68 => i:9d4fcabf Nodo: Get BP -> InicioAccion: getBloodPressure 27/05/2013 19:31:18.69 => i:9d4fcabf Nodo: Get BP -> InicioReloj: clk4 27/05/2013 19:31:18.71 => i:9d4fcabf Nodo: Get Temp -> InicioAccion: getTemperature 27/05/2013 19:31:18.72 => i:9d4fcabf Nodo: Get Temp -> InicioReloj: clk3 27/05/2013 19:31:18.83 => i:8229056d Fin Accion: FirstTriageResult Res: ACUTE 27/05/2013 19:31:18.88 => i:8229056d Nodo: Get BP -> InicioAccion: getBloodPressure 27/05/2013 19:31:18.90 => i:8229056d Nodo: Get BP -> InicioReloj: clk4 27/05/2013 19:31:18.91 => i:8229056d Nodo: Get Temp -> InicioAccion: getTemperature 27/05/2013 19:31:18.93 => i:8229056d Nodo: Get Temp -> InicioReloj: clk3 27/05/2013 19:31:19.47 => i:1a5c92f6 Fin Accion: getTemperatureResult Res: FEVER 27/05/2013 19:31:20.19 => i:aa3380da Fin Accion: getTemperatureResult Res: OK 27/05/2013 19:31:20.27 => i:aa3380da Nodo: TEMP OK -> InicioAccion: 27/05/2013 19:31:20.38 => i:aa3380da Fin Accion: getBloodPressureResult Res: OK 27/05/2013 19:31:20.40 => i:aa3380da Nodo: BP OK -> InicioAccion: 27/05/2013 19:31:20.41 => i:aa3380da Nodo: Quality Test -> InicioAccion: QualityTest …
- 6. Uses of Process Mining Using workflows to describe Clinical Pathways/Care processes Facilitate the praxis of health professionals Improvement of quality of care Unify criteria Help the administrative management of clinical processes Process Mining can infer the real deployment of processes Computer Aided Design of Clinical Pathways Clinical Pathways tracing Detect bottlenecks Cost effectiveness study of care processes Measure adherence of Clinical Pathways instances Processes deployment support Individualized behavior modeling
- 8. Spaghetti Effect This is not useful at all!
- 9. Spaghetti Effect This is not useful at all!
- 10. Recommended practices: Using Space-Efficient Structures Petri net Determinate Finite Automation DFA
- 11. Recommended practices: Time Abstractions Use of discrete time variables to reduce Spaghetti effect
- 12. Recommended practices: Rendering Algorithms Based on different profiles/priorities
- 13. Recommended practices: Clustering Showing the different paths and frequencies to know more about the problem and find a better solution
- 14. Recommended practices: Workflow layout and navigation
- 15. Discussion/Conclusions Process Mining can be used to support clinical pathways design and traceability The Spaguetti Effect is a well known problem that affects critically the application of process mining to health The use of such recommended practices allows a mitigation of the spaguetti effect making process mining techniques usable in clinical scenarios The application of Interactive Pattern Recognition Technologies in the continuous following of patients with diabetes could be a very interesting alternative to manual creation of Clinical Pathways (or Life Assistance Protocols) offering not only a way to support the design of those protocols but also support the deployment of these protocols, offering a way to continuously improve them and tracking the patient throughout all the process.
- 16. Innovaciones Tecnologías para la Salud y el Bienestar Vicente Traver @vtraver vtraver@itaca.upv.es MIlano, 26th August 2015 Diabetes care related process modelling using Process Mining techniques. Lessons learned in the application of Interactive Pattern Recognition: coping with the Spaghetti Effect Questions ?
Editor's Notes
- #3: Diabetes is one of the metabolic disorders with more growth expectations in next decades. The worldwide prevalence of Diabetes was estimated around 8.3% among adults between 20 and 80 years[2] and it is expected that the number of patients with diabetes will continue increasing dramatically in the following decades[2][3]. The literature points to a correct self-management, to an appropriate treatment and to an adequate healthy lifestyle as a way to dramatically improve the quality of life of patients with diabetes. The implementation of a holistic diabetes care system, using rising information technologies for deploying cares based on the thesis of the Evidence-Based Medicine can be a effective solution to provide an adequate and continuous care to patients. However, the design and deployment of computer readable careflows is not a easy task. A correct self-management of the illness combined with a proper medical treatment and an adequate healthy lifestyle can lead to a dramatic reduction in the mortality rates and in the reduction of co-morbidities and complications associated to the illness, allowing the patient to have a normal life[4]
- #9: Search a needle in a haystack
- #10: Search a needle in a haystack
- #11: This implies better understanding/readability vs accuracy