In a significant stride towards enhancing breast cancer detection, a team of researchers led by Omneya Attallah from the Department of Electronics and Communications Engineering at the Arab Academy for Science, Technology and Maritime Transport has developed a novel computer-aided diagnosis (CAD) system named Thermo-CAD. This system leverages infrared thermography and advanced deep learning techniques to improve early breast cancer detection, a critical factor in increasing survival rates. The study, published in the journal ‘Scientific Reports’ (translated from Arabic), offers promising avenues for medical diagnostics, with potential implications for maritime sectors, particularly in remote and resource-constrained environments.
Breast cancer remains a prevalent concern among women globally, with early detection being key to effective treatment. Traditional mammography, while widely used, has limitations such as accessibility issues and patient discomfort. Thermo-CAD aims to address these challenges by utilizing thermal imaging, a non-invasive and radiation-free method that captures temperature maps of the breast to identify potential tumors based on thermal irregularities.
The Thermo-CAD system employs multiple convolutional neural networks (CNNs) to enhance diagnostic accuracy. According to Attallah, “The integration of multiple deep features and the use of feature transformation and selection methods, such as non-negative matrix factorization and Relief-F, significantly reduce classification complexity and improve reliability.” This advanced approach allows the system to analyze thermal images with high precision, providing radiologists with valuable assistance in early-stage cancer detection.
The system was evaluated using two datasets: the DMR-IR (Database for Mastology Research Infrared Images) and a novel thermography dataset. Thermo-CAD demonstrated outstanding performance on the DMR-IR dataset, achieving 100% accuracy in distinguishing between normal and abnormal breast tissues using CSVM and MGSVM classifiers. However, its ability to differentiate between benign and malignant cases was lower, with an accuracy of 79.3% using the CSVM classifier. Despite this, the system shows great promise for early-stage cancer detection, particularly in areas where access to traditional diagnostic tools is limited.
For maritime professionals, the implications of this research are noteworthy. Ships operating in remote regions often face challenges in accessing medical facilities equipped with advanced diagnostic tools. Thermo-CAD’s portability and non-invasive nature make it an ideal solution for onboard medical care, ensuring that crew members can receive timely and accurate diagnoses. Additionally, the system’s potential to reduce the need for invasive procedures aligns with the maritime industry’s focus on crew welfare and safety.
The commercial opportunities are equally compelling. The development of Thermo-CAD opens doors for maritime medical service providers to integrate advanced diagnostic technologies into their offerings. This not only enhances the quality of healthcare services but also positions these providers as leaders in innovative medical solutions. Furthermore, the system’s ability to operate in resource-constrained environments makes it a valuable asset for humanitarian missions and disaster relief efforts, where access to medical facilities is often limited.
In summary, Thermo-CAD represents a significant advancement in breast cancer detection, offering a non-invasive, radiation-free alternative to traditional mammography. Its potential to improve early-stage cancer detection, particularly in remote and resource-constrained environments, makes it a valuable tool for maritime professionals. As Attallah notes, “Thermo-CAD’s outstanding performance on the DMR-IR dataset highlights its potential to revolutionize breast cancer diagnostics.” The commercial opportunities for maritime sectors are substantial, paving the way for enhanced medical services and innovative solutions in the industry.