A team of Chinese scientists have developed an artificial intelligence (AI) tool that predicts hard-to-identify tumor origins with an accuracy rivaling or even surpassing human pathologists.

Cancer of unknown primary (CUP) site poses challenges to clinicians due to its elusive nature. The CUP, which accounts for 3 to 5 percent of all cancers diagnosed in humans, tends to be malignant, with only 20 percent of CUP patients achieving a median survival of 10 months.

Led by researchers from Tianjin Medical University and the First Affiliated Hospital of Zhengzhou University, a deep-learning method was developed for tumor origin differentiation, which was trained via cytological images from 57,220 cases at four Chinese hospitals.

The tool called TORCH can identify malignant tumors in fluids accumulated in the chest and abdomen and predict their origins, according to a study published in the journal Nature Medicine.

TORCH achieved a prediction accuracy of 82.6 percent, marking a significant improvement compared with that of four human pathologists, according to the study.

Furthermore, the treatment protocol given in conforming with TORCH-predicted origins resulted in an overall survival of 27 months as against 17 months for those who were administrated discordant treatment.

The study highlighted the AI system's potential as a valuable ancillary tool in clinical practice, although further validation in randomized trials is required, said the researchers.

Source: Xinhua