Human disease prediction is the process of estimating the likelihood that a patient would develop a
disease after analysing the constellations of their symptoms. Keeping track of a patient's health
information and status during the initial assessment can help doctors treat a patient's condition
successfully. Patients would receive a more streamlined and rapid course of therapy as a result of this
analysis in the medical field. This study aims for the detection severity of multiple diseases in patients
using machine learning (ML) models. Google Colab – python 3.0 with libraries from Sklearn, Numpy,
Pandas, Matplotlib was the platform that was used to classify the data. There are 222 records of patients
with varying health problems. Data from 2018 – 2023, among which 104 were female and 118 were
male around the age group of 7-81 years were collected from a private hospital in Mysuru, Karnataka.
The dataset included information on the patient's age, gender, blood pressure, respiratory rate, pulse,haemoglobin, total leucocyte count, red blood cell count, platelet count, etc. In this study, the ability to
categorise the severity of multiple diseases using six ML models - Logistic Regression (LR), K-Nearest
Neighbour (K-NN), Support Vector Machine-Radial basis function Kernel (SVM-rbf), Gaussian,
Decision Tree (DT) and Random Forest (RF) - was studied. DT model attained the best accuracy for
detecting severity, with 99% accuracy, while RF model obtained 98%, Gaussian model obtained 80%,
SVM model obtained 64% and LR model and the KNN obtained 62% accuracy.
Keywords: Human Health, Machine Learning, Classification, Decision Tree, Accuracy
Publication date: 01/02/2026
https://ijbpas.com/pdf/2026/February/MS_IJBPAS_2026_9513.pdf
Download PDF
https://doi.org/10.31032/IJBPAS/2026/15.2.9513
