HELSINKI, Feb. 27 (Xinhua) -- The University of Jyvaskyla of Finland has developed a new artificial intelligence (AI)-based tool to automatically identify colorectal cancer from tissue sample analysis, the university announced on Thursday.

It said that the neural network model developed in the study has surpassed all its predecessors in the classification of tissue samples.

Based on the study, the new model can identify all tissue categories relevant for colorectal cancer identification, with an accuracy of 96.74 percent, said Fabi Prezja, the researcher responsible for designing the method.

Conventionally, tissue analysis requires a pathologist to look through slides of scanned tissue samples, and mark point by point where the cancerous and related tissues are visible.

However, the new AI-based tool analyses a sample and automatically highlights areas containing different tissue categories. Its accuracy has the potential to significantly ease the workload of histopathologists, potentially resulting in faster diagnoses, prognoses and clinical insights, explained the university.

The research team has made the AI-based tool freely available, in order to encourage research collaboration.

"The free availability aims to accelerate future advances by encouraging scientists, developers and researchers worldwide to continue developing the tool and finding new applications for it," said Prezja.

Nevertheless, the research team pointed out that despite promising results, the introduction of AI tools to the clinical setting must be done gradually and with caution.

The AI-based model has been developed by researchers at the University of Jyvaskyla, in collaboration with the University of Turku's Institute of Biomedicine, University of Helsinki and Nova Hospital of Central Finland. Enditem

Related

Go to Forum >>0 Comment(s)