Healthcare Risk Analytics: An Interactive Dashboard and Machine Learning Model for Patient Risk Prediction
A huge amount of information about patients is collected in hospitals, however, the data is stored separately in tables, and the risk assessment is done manually. In this paper we introduce the Healthcare Risk Analytics – a light-weight web-application developed with Python and Streamlit that combines data about patients, diagnoses, outcomes, and laboratories results in a single interactive dashboard and also implements a supervised machine learning classifier to determine whether a patient is High Risk or Low Risk based on his/her age, length of stay, treatment cost, and abnormal laboratory count with probability output. The dashboard includes filter options, KPI, cost and length of stay analysis per diagnoses, distribution of laboratory results, patient table for search purposes, and a page with predictions. The system was evaluated through such metrics as accuracy, precision, recall, F1-score and also tested on a set of functional test cases. The paper demonstrates that a small team of a hospital can easily analyze its records and assess the risk for each patient using free tools and a small dataset. KEYWORDS Healthcare Analytics, Predicting Patient Risk, Machine Learning, Streamlit, Dashboard, Power BI.
Authors
- Paras Sharma (ORCID: https://orcid.org/0000-0003-0107-2666)
- Paarth kumar
- Lakshay Lakshay
Institutions
- Jagan Institute of Management Studies (IN)
Publication Details
- Journal
- International Journal of Creative and Open Research in Engineering and Management
- Published
- 2026-11-19
- DOI
- https://doi.org/10.55041/ijcope.v2i11.115
- Primary Topic
- Artificial Intelligence in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00