Open Access

Predictive diagnostics for early identification of cardiovascular disease: a machine learning approach

Pooja Bagane

Pooja Bagane

Department of Computer Science and Engineering, Symbiosis Institute of Technology – Pune Campus, Symbiosis International (Deemed University)Pune, India
Search for this author on
Sciendo|Google Scholar
Bagane, Pooja
, 
Moksh Oswal

Moksh Oswal

Department of Computer Science and Engineering, Symbiosis Institute of Technology – Pune Campus, Symbiosis International (Deemed University)Pune, India
Search for this author on
Sciendo|Google Scholar
Oswal, Moksh
, 
Sachin Mhetre

Sachin Mhetre

Department of Computer Science and Engineering, Symbiosis Institute of Technology – Pune Campus, Symbiosis International (Deemed University)Pune, India
Search for this author on
Sciendo|Google Scholar
Mhetre, Sachin
, 
Prabhat Shankar

Prabhat Shankar

Department of Computer Science and Engineering, Symbiosis Institute of Technology – Pune Campus, Symbiosis International (Deemed University)Pune, India
Search for this author on
Sciendo|Google Scholar
Shankar, Prabhat
, 
Praneet Mahendrakar

Praneet Mahendrakar

Department of Computer Science and Engineering, Symbiosis Institute of Technology – Pune Campus, Symbiosis International (Deemed University)Pune, India
Search for this author on
Sciendo|Google Scholar
Mahendrakar, Praneet
 and 
Obsa Amenu Jebessa

Obsa Amenu Jebessa

Department of Information Technology, Jimma Institute of TechnologyJimma, Ethiopia
Search for this author on
Sciendo|Google Scholar
Jebessa, Obsa Amenu
  
May 16, 2025
Predictive diagnostics for early identification of cardiovascular disease: a machine learning approach's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Article Category: Research Article
Published Online: May 16, 2025
Received: Jan 12, 2025
DOI: https://doi.org/10.2478/ijssis-2025-0021
Keywords
© 2025 Pooja Bagane et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1:

Block diagram for predictive diagnostics for early identification of cardiovascular disease. KNN, K-nearest neighbors; MLP, multilayer perceptron; PCA, principal component analysis.
Block diagram for predictive diagnostics for early identification of cardiovascular disease. KNN, K-nearest neighbors; MLP, multilayer perceptron; PCA, principal component analysis.

Figure 2:

Correlation heatmap illustrating the interrelationship between features in the Cleveland dataset.
Correlation heatmap illustrating the interrelationship between features in the Cleveland dataset.

Figure 3:

Correlation heatmap illustrating the interrelationship between features in the Statlog dataset.
Correlation heatmap illustrating the interrelationship between features in the Statlog dataset.

Figure 4:

Count plot (age-count) on the Cleveland dataset.
Count plot (age-count) on the Cleveland dataset.

Figure 5:

Count plot (age-count) on the Statlog dataset.
Count plot (age-count) on the Statlog dataset.

Figure 6:

Thalach distribution plot on the Cleveland dataset.
Thalach distribution plot on the Cleveland dataset.

Figure 7:

Thalach distribution plot on the Statlog dataset.
Thalach distribution plot on the Statlog dataset.

Figure 8:

MLP predictions confusion matrix on the Statlog dataset. MLP, multilayer perceptron.
MLP predictions confusion matrix on the Statlog dataset. MLP, multilayer perceptron.

Figure 9:

MLP predictions confusion matrix on the Cleveland dataset. MLP, multilayer perceptron.
MLP predictions confusion matrix on the Cleveland dataset. MLP, multilayer perceptron.

Figure 10:

KNN predictions confusion matrix on the Statlog dataset. KNN, K-nearest neighbors.
KNN predictions confusion matrix on the Statlog dataset. KNN, K-nearest neighbors.

Figure 11:

KNN predictions confusion matrix on the Cleveland dataset. KNN, K-nearest neighbors.
KNN predictions confusion matrix on the Cleveland dataset. KNN, K-nearest neighbors.

Figure 12:

LR predictions confusion matrix on the Statlog dataset. LR, logistic regression.
LR predictions confusion matrix on the Statlog dataset. LR, logistic regression.

Figure 13:

LR predictions confusion matrix on the Cleveland dataset. LR, logistic regression.
LR predictions confusion matrix on the Cleveland dataset. LR, logistic regression.

Comparative analysis of models on the Statlog dataset

Method Accuracy (%) Precision (%) F1-score (%) Recall (%)
MLP 94.81 97.33 95.42 93.59
LR 92.6 86 89 87
KNN 78 85 90 87

Dataset description

No. Dataset description Ranges
1 Age (years) 29–79
2 Sex Male (1), female (0)
3 Types of chest pain (cp) Typical angina (0), atypical angina (1), non-anginal pain (2), asymptomatic (3)
4 Resting blood pressure (trestbps) 94–200 (mmHg)
5 Serum cholesterol (chol) 126–564 (mg/dL)
6 Fasting blood pressure (fbs) False (0), true (1) (mg/dL)
7 Resting electrocardiographic results (restecg) Normal (0), having ST-T wave abnormality (1) probable or definite left ventricular hypertrophy (2)
8 Maximum heart rate achieved (thalach) 71–202
9 Exercise-induced angina (exang) No (0), yes (1)
10 ST depression induced by exercises relative to rest (oldpeak) 0–6.2
11 Slope of the peak exercises ST segment (slope) Upsloping (0), flat (1), downsloping (2)
12 Number of major vessels colored by fluoroscopy (ca) 0–3
13 Thalassemia (thal) Normal (l), fixed defect (2), reversible defect (3)
14 Target class (target) No (0), yes (1)

Comparative analysis of models on the Cleveland dataset

Method Accuracy (%) Precision (%) F1-score (%) Recall (%)
MLP 94.39 95.73 94.86 94.01
LR 85 85 89 87
KNN 92.8 87 91 89
Language:
English
Publication timeframe:
1 times per year
Journal Subjects:
Engineering, Introductions and Overviews, Engineering, other
Journal RSS Feed