Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Leveraging AI for Automatic Classification of PCOS Using Ultrasound Imaging

Atharva Divekar, Atharva Sonawane

TL;DR

This work tackles automatic PCOS diagnosis from ultrasound frames by leveraging transfer learning with InceptionV3 and a custom classifier head. The PCOSGen dataset (4,668 images) enables binary healthy/unhealthy classification, achieving strong validation performance (accuracy ~90.5%, recall ~97.2%, precision ~90.0%) and providing interpretability through LIME explanations. Extensive model experimentation shows InceptionV3 outperforms ResNet101, EfficientNet B7, and RadImagenet variants, supporting its selection for this task. The study demonstrates the feasibility of interpretable, AI-driven PCOS diagnostics from ultrasound data and outlines practical directions for broader validation and clinical deployment.

Abstract

The AUTO-PCOS Classification Challenge seeks to advance the diagnostic capabilities of artificial intelligence (AI) in identifying Polycystic Ovary Syndrome (PCOS) through automated classification of healthy and unhealthy ultrasound frames. This report outlines our methodology for building a robust AI pipeline utilizing transfer learning with the InceptionV3 architecture to achieve high accuracy in binary classification. Preprocessing steps ensured the dataset was optimized for training, validation, and testing, while interpretability methods like LIME and saliency maps provided valuable insights into the model's decision-making. Our approach achieved an accuracy of 90.52%, with precision, recall, and F1-score metrics exceeding 90% on validation data, demonstrating its efficacy. The project underscores the transformative potential of AI in healthcare, particularly in addressing diagnostic challenges like PCOS. Key findings, challenges, and recommendations for future enhancements are discussed, highlighting the pathway for creating reliable, interpretable, and scalable AI-driven medical diagnostic tools.

Leveraging AI for Automatic Classification of PCOS Using Ultrasound Imaging

TL;DR

This work tackles automatic PCOS diagnosis from ultrasound frames by leveraging transfer learning with InceptionV3 and a custom classifier head. The PCOSGen dataset (4,668 images) enables binary healthy/unhealthy classification, achieving strong validation performance (accuracy ~90.5%, recall ~97.2%, precision ~90.0%) and providing interpretability through LIME explanations. Extensive model experimentation shows InceptionV3 outperforms ResNet101, EfficientNet B7, and RadImagenet variants, supporting its selection for this task. The study demonstrates the feasibility of interpretable, AI-driven PCOS diagnostics from ultrasound data and outlines practical directions for broader validation and clinical deployment.

Abstract

The AUTO-PCOS Classification Challenge seeks to advance the diagnostic capabilities of artificial intelligence (AI) in identifying Polycystic Ovary Syndrome (PCOS) through automated classification of healthy and unhealthy ultrasound frames. This report outlines our methodology for building a robust AI pipeline utilizing transfer learning with the InceptionV3 architecture to achieve high accuracy in binary classification. Preprocessing steps ensured the dataset was optimized for training, validation, and testing, while interpretability methods like LIME and saliency maps provided valuable insights into the model's decision-making. Our approach achieved an accuracy of 90.52%, with precision, recall, and F1-score metrics exceeding 90% on validation data, demonstrating its efficacy. The project underscores the transformative potential of AI in healthcare, particularly in addressing diagnostic challenges like PCOS. Key findings, challenges, and recommendations for future enhancements are discussed, highlighting the pathway for creating reliable, interpretable, and scalable AI-driven medical diagnostic tools.
Paper Structure (20 sections, 3 figures, 2 tables)

This paper contains 20 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Key Characteristics of the Dataset
  3. Methods
  4. Image Preprocessing
  5. Data Preparation
  6. Base Model Initialization
  7. Model Architecture
  8. Model Compilation
  9. Model Training
  10. Monitoring Training Progress
  11. Model Evaluation
  12. Model Experimentation
  13. Results
  14. Model Performance Metrics
  15. Comparion with Baseline Models
  16. ...and 5 more sections

Figures (3)

  • Figure 3: Deep Learning Pipeline for PCOS Image Classification. This figure illustrates the key stages of the image classification pipeline, including preprocessing, data preparation, model training, and evaluation.
  • Figure 4: Comparison of LIME explanations for a correctly classified validation image (left) and test image (right). In both instances, the model accurately predicted Healthy, with LIME highlighting key regions which model focus on.
  • Figure 5: Comparison of LIME explanations for a correctly classified validation image (left) and test image (right). In both instances, the model accurately predicted Unhealthy, with LIME highlighting key regions which model focus on.