In this paper we present a method and a tool to acquire data of outpatients suffering from Heart Failure and to populate a database so that it is suitable for the supervised training of machine learning techniques. In our studies we had to train an artificial intelligence-based system to recognize different severity and to predict worsening of heart failure patients, using as input various parameters acquirable during outpatient visits. We have therefore developed a tool that would allow the cardiologist to populate a ”supervised database” suitable for machine learning during his regular outpatient consultations. The idea comes from the fact that in literature there are few databases of this type and they are not scalable in our case. The tool includes a management part for the patient demographics, a part for the data acquisition, a part for displaying the follow-ups of a patient, a part of artificial intelligence to provide the smart output, and a part called ”score based prognosis” in which we have computerized some prognostic models known in the literature as the SHFM, the CHARM, the EFFECT, and ADHERE.
[1]
W John Boscardin,et al.
Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis.
,
2005,
JAMA.
[2]
G. Eknoyan,et al.
National Kidney Foundation Practice Guidelines for Chronic Kidney Disease: Evaluation, Classification, and Stratification
,
2003,
Annals of Internal Medicine.
[3]
Peter C. Austin,et al.
Predicting mortality among patients hospitalized for heart failure: derivation and validation of a clinical model.
,
2003,
JAMA.
[4]
Gabriele Guidi,et al.
Heart Failure analysis Dashboard for patient's remote monitoring combining multiple artificial intelligence technologies
,
2012,
2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.