AI Model Achieves 90% Accuracy in Detecting Lymphatic Cancer

Researchers at Chalmers University of Technology have developed an artificial intelligence (AI) model that can detect lymphatic cancer in 90% of cases, marking a significant advancement in medical image analysis. Named Lars (Lymphoma Artificial Reader System), this deep learning-based AI system analyzes images from positron emission tomography (PET) and computed tomography (CT) to identify signs of lymphoma, a cancer of the lymphatic system.

This breakthrough is the result of one of the largest studies of its kind, involving over 17,000 images from more than 5,000 lymphoma patients. The study, which examined image archives spanning more than a decade, compared patients’ final diagnoses with their scans before and after treatment to train the AI to detect lymphoma.

The development of AI-assisted image analysis for medical conditions aims to support radiologists by reducing workload, offering a second opinion, and prioritizing patient treatment. Moreover, it promotes equality in healthcare by ensuring patients have access to consistent expertise and timely image reviews, regardless of their location. This is particularly beneficial for rare diseases where radiologists may not frequently encounter relevant images.

The AI model uses supervised training, where it is shown images and learns to assess whether a patient has lymphoma based on the patterns and features within those images. This method allows the AI to improve its diagnostic accuracy over time without being programmed with specific instructions on what to look for in the images.

Despite the high accuracy rate and potential benefits, further validation is required before Lars can be integrated into clinical practice. The study’s findings have been made available to the scientific community to encourage further research and development. This initiative could significantly impact the diagnosis and treatment of lymphoma and other medical conditions in the future.

Source: Medicalxpress

More information: Ida Häggström et al, Deep learning for [¹⁸F]fluorodeoxyglucose-PET-CT classification in patients with lymphoma: a dual-centre retrospective analysis, The Lancet Digital Health (2023). DOI: 10.1016/S2589-7500(23)00203-0

---

Like this article?  Keep up to date with AI news, apps, tools and get tips and tricks on how to improve with AI.  Sign up to our Free AI Newsletter

Also, come check out our free AI training portal and community of business owners, entrepreneurs, executives and creators. Level up your business with AI ! New courses added weekly. 

You can also follow us on X