Thyroid Disease Detection Using Machine Learning Approach

EasyChair Preprint no. 10684

Abstract

Thyroid disorders are prevalent worldwide and
can significantly impact an individual's health and well
being. The accurate detection and diagnosis of thyroid
diseases are crucial for effective management and treatment.
 The most common thyroid
hypothyroidism. Hypo- means deficient or under (active), so
hypothyroidism is a condition in which the thyroid gland is
underperforming or producing too little thyroid hormone.
Recognizing the symptoms of hypothyroidism is extremely
important. The proposed program leverages a diverse
a dataset comprising various thyroid-related parameters,
including patient demographics, medical history and
laboratory test results. By harnessing the power of machine
learning algorithms, the program learns intricate
and predicts accuracy accordingly. The program employs
several machine learning techniques to build a robust and
reliable thyroid disease detection model, including feature
extraction, feature selection, and classification algorithms
We take the assistance of RandomForestClassifier and
StandardScaler Through an iterative training process, the
the program optimizes the model's performance by minimizing
false positives and false negatives, ensuring accurate
predictions and reducing the likelihood of mi
program's performance is compared against existing
diagnostic methods, including clinical guidelines and expert
interpretations of medical professionals, to validate its
efficacy and potential for clinical adoption.The results of the
the evaluation demonstrates that the machine learning
thyroid detection program achieves superior performance in
terms of accuracy and efficiency compared to traditional
diagnostic approaches.

Keyphrases: extraction, Hypothyroidism, RandomForestClassifier, Thyroid

@Booklet{EasyChair:10684,
  author = {V Viswanatha and A.C Ramachandra and Rahul R Sankar and Chinmay R Sanjay Mahulika},
  title = {Thyroid Disease Detection Using Machine Learning Approach},
  howpublished = {EasyChair Preprint no. 10684},

  year = {EasyChair, 2023}}