CACTUS: An Open Dataset and Framework for Automated Cardiac Assessment and Classification of Ultrasound Images Using Deep Transfer Learning

TL;DR

This work tackles the lack of publicly available graded cardiac US data by introducing CACTUS, a phantom-derived dataset spanning multiple heart views with expert-quality grades. It presents a transfer-learning framework built on a shared encoder (ResNet18) with two heads for simultaneous cardiac-view classification and image-quality grading, achieving up to 99.43% validation accuracy for view classification and 0.3067 MSE for grading. The study demonstrates TL advantages over MTL, shows strong adaptability to new views (e.g., six-class extension with PSMV achieving 99.99% accuracy), and confirms practical viability through real-time scans and expert feedback (average 8/10). Overall, CACTUS and the TL-based framework offer a scalable path toward automated, reliable echocardiography support with potential for real-time clinical integration and ongoing dataset expansion.

Abstract

Cardiac ultrasound (US) scanning is a commonly used techniques in cardiology to diagnose the health of the heart and its proper functioning. Therefore, it is necessary to consider ways to automate these tasks and assist medical professionals in classifying and assessing cardiac US images. Machine learning (ML) techniques are regarded as a prominent solution due to their success in numerous applications aimed at enhancing the medical field, including addressing the shortage of echography technicians. However, the limited availability of medical data presents a significant barrier to applying ML in cardiology, particularly regarding US images of the heart. This paper addresses this challenge by introducing the first open graded dataset for Cardiac Assessment and ClassificaTion of UltraSound (CACTUS), which is available online. This dataset contains images obtained from scanning a CAE Blue Phantom and representing various heart views and different quality levels, exceeding the conventional cardiac views typically found in the literature. Additionally, the paper introduces a Deep Learning (DL) framework consisting of two main components. The first component classifies cardiac US images based on the heart view using a Convolutional Neural Network (CNN). The second component uses Transfer Learning (TL) to fine-tune the knowledge from the first component and create a model for grading and assessing cardiac images. The framework demonstrates high performance in both classification and grading, achieving up to 99.43% accuracy and as low as 0.3067 error, respectively. To showcase its robustness, the framework is further fine-tuned using new images representing additional cardiac views and compared to several other state-of-the-art architectures. The framework's outcomes and performance in handling real-time scans were also assessed using a questionnaire answered by cardiac experts.

Figures (21)

• Figure 1: Different images of the apical four chamber cardiac view
• Figure 2: Graded cardiac views
• Figure 3: An illustration of the CAE Blue Phantom scanning setup
• Figure 4: Data acquisition process
• Figure 5: Subcostal four chamber cardiac view