I’m Md. Abu Sama

The main purpose of this research is to develop an efficient machine learning-based system that can predict the likelihood of diabetes at an early stage using health-related parameters. Early prediction of diabetes enables timely medical intervention and lifestyle adjustments, which can significantly reduce the risk of severe complications. This study aims to compare the performance of multiple machine learning classifiers to identify the most accurate and reliable model for early diabetes detection.

Data Collection :

The dataset used in this research was obtained from publicly available and verified medical repositories, such as the PIMA Indians Diabetes Database from the UCI Machine Learning Repository from Kaggle. The dataset contains medical and demographic information of female patients aged 21 years and older of Pima Indian heritage. Each record includes several health attributes such as:

• Number of pregnancies
• Glucose concentration
• Blood pressure
• Skin thickness
• Insulin level
• Body mass index (BMI)
• Diabetes pedigree function
• Age
• Outcome (indicating whether the patient has diabetes or not)

Research Methodology :

The research methodology involves the following steps:

1. Data Preprocessing: Cleaning and preparing the dataset for analysis.
2. Feature Selection: Identifying the most relevant features for diabetes prediction.
3. Model Training: Training multiple machine learning models on the dataset.
4. Model Evaluation: Comparing the performance of different models using appropriate metrics.

Results :

Several supervised machine learning classifiers were trained and tested, including Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). After evaluation, the Support Vector Machine (SVM) Classifier achieved the highest accuracy of approximately 92%, outperforming other models in terms of precision, recall, and F1-score.

Conclusion :

The expected outcome of this research is to identify a machine learning model that can accurately predict the likelihood of diabetes in patients at an early stage. This model can potentially be integrated into clinical decision support systems to assist healthcare professionals in identifying at-risk individuals and recommending preventive measures.

Author :

Md. Abu Sama

The Main Purpose of this research is to find out a new way of taking decisions to classify a risky pregnant mother, on the absence of doctor.

Data Collection :

Collect Raw data from various hospitals of Rangamati, Khagrachori and Bandarban, which provide maternal care for pregnant mothers. And transform those raw page data into Machine analysis fessible data.

Research Methodology :

The research methodology involves the following steps:

1. Data Preprocessing: Cleaning and preparing the dataset for analysis.
2. Feature Selection: Identifying the most relevant features for risky pregnant mother classification.
3. Model Training: Training multiple machine learning models on the dataset.
4. Model Evaluation: Comparing the performance of different models using appropriate metrics.

Results :

Several supervised machine learning classifiers were trained and tested, including Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). After evaluation, the Random Forest Classifier achieved the highest accuracy of approximately 89%, outperforming other models in terms of precision, recall, and F1-score.

Conclusion :

The expected outcome of this research is to identify a machine learning model that can accurately classify risky pregnant mothers. This model can potentially be integrated into healthcare systems to assist healthcare professionals in identifying at-risk individuals and recommending preventive measures.

Author :

Md. Abu Sama