A new artificial intelligence model developed by Harvard Medical School has shown remarkably high accuracy in detecting cancer.

Called CHIEF (Clinical Histopathology Imaging Evaluation Foundation), the AI model was able to detect cancer with 96% accuracy across 19 different types of the disease in the study conducted by senior author Kun-Hsing Yu and colleagues. The study’s findings were published in Nature on Sept. 4.

“Our ambition was to create a nimble, versatile ChatGPT-like AI platform that can perform a broad range of cancer evaluation tasks… Our model turned out to be very useful across multiple tasks related to cancer detection, prognosis, and treatment response across multiple cancers,” said Yu, per the Harvard Medical School website.

CLICK HERE TO GET THE DALLAS EXPRESS APP

CHIEF was trained on a vast multimodal dataset that included 15 million unlabeled images and “60,530 whole-slide images spanning 19 anatomical sites,” says the study. In total, 44 terabytes of high-resolution images were leveraged to train CHIEF. This enabled the software to identify “microscopic representations useful for cancer cell detection, tumor origin identification, molecular profile characterization and prognostic prediction.”

CHIEF is only expected to improve with time. Researchers intend to train the model on images of rare diseases, non-cancerous conditions, and pre-malignant tissues, all with the goal of enhancing accuracy.

This is not the first time artificial intelligence has been leveraged to help identify cancer. Earlier this year, The Dallas Express reported that a model developed by researchers at Duke University in Durham, North Carolina, was able to help predict future breast cancer risk in patients up to five years earlier than a previously released model. Known as AsymMirai, the model was a simplified enhancement of the prior version called simply Mirai.

“If validated further and deployed widely, our approach, and approaches similar to ours, could identify early on cancer patients who may benefit from experimental treatments targeting certain molecular variations, a capability that is not uniformly available across the world,” Yu said of CHIEF.