Authors :
Sahil Kumar Suman; Natasha Sharma; Udeshna Saikia; Dhiti; Rahul Chauhan; Nandini Singh
Volume/Issue :
Volume 8 - 2023, Issue 11 - November
Google Scholar :
https://tinyurl.com/y6jm754m
Scribd :
https://tinyurl.com/4r89exwv
DOI :
https://doi.org/10.5281/zenodo.10319575
Abstract :
Diabetes has been recorded as a serious glob-
al health issue today. It's a long-term metabolic disease
that takes place when blood glucose levels elevate in the
human body. Early and accurate diabetes diagnosis is
essential for managing the condition precisely and will
prevent complications quickly. This count proposes a
comprehensive and effective machine-learning method
for detecting and treating diabetes. The dataset that was
used contains many clinical and demographic variables
such as age, BMI, family history and various blood test
results. To identifythe most relevant variables, the tech-
nique prioritizes the data to control for missing values
and to normalize features. The next stepis to go through a
strict feature selection process. For the trainingand vali-
dation of the model, SVM, RFM, Logistic Regression,
and Support Vector Machines (SVM) are just a few of
the machine learning algorithms that are employed. The
performance of each of these algorithms is checked using
metrics like accuracy, redundancy, uniqueness, and re-
ceiver operating characteristic (ROC) curve area. An en-
semble perspective is also explored to combine thebenefits
of multiple models and increase overall predicting power.
The recommended model is tested on various test da-
tasets for assessment purposes of its generalizability. The
main purpose of the project is to create a robust and
trustworthy diabetes detection tool that can be used in
clinical settings to aid medical professionals with ad-
vanced diagnosis and individualized treatment planning.
The results demonstrate growing performance and the
potential for machine learning to increase diabetes de-
tection accuracy. The importance of the proposed model
to subtle patterns in different patient data sets suggests
that it could apply to a large range of demographics.
This work lays the root level for future analysis into en-
hancing and expanding the capabilities of diabetesdetec-
tion models, which will advance ongoing efforts to apply
machine learning to healthcare applications.
Keywords :
Diabetes, Early Diagnosis, Machine Learning, Ac-Curacy, Healthcare Applications.
Diabetes has been recorded as a serious glob-
al health issue today. It's a long-term metabolic disease
that takes place when blood glucose levels elevate in the
human body. Early and accurate diabetes diagnosis is
essential for managing the condition precisely and will
prevent complications quickly. This count proposes a
comprehensive and effective machine-learning method
for detecting and treating diabetes. The dataset that was
used contains many clinical and demographic variables
such as age, BMI, family history and various blood test
results. To identifythe most relevant variables, the tech-
nique prioritizes the data to control for missing values
and to normalize features. The next stepis to go through a
strict feature selection process. For the trainingand vali-
dation of the model, SVM, RFM, Logistic Regression,
and Support Vector Machines (SVM) are just a few of
the machine learning algorithms that are employed. The
performance of each of these algorithms is checked using
metrics like accuracy, redundancy, uniqueness, and re-
ceiver operating characteristic (ROC) curve area. An en-
semble perspective is also explored to combine thebenefits
of multiple models and increase overall predicting power.
The recommended model is tested on various test da-
tasets for assessment purposes of its generalizability. The
main purpose of the project is to create a robust and
trustworthy diabetes detection tool that can be used in
clinical settings to aid medical professionals with ad-
vanced diagnosis and individualized treatment planning.
The results demonstrate growing performance and the
potential for machine learning to increase diabetes de-
tection accuracy. The importance of the proposed model
to subtle patterns in different patient data sets suggests
that it could apply to a large range of demographics.
This work lays the root level for future analysis into en-
hancing and expanding the capabilities of diabetesdetec-
tion models, which will advance ongoing efforts to apply
machine learning to healthcare applications.
Keywords :
Diabetes, Early Diagnosis, Machine Learning, Ac-Curacy, Healthcare Applications.
