A new artificial intelligence tool will help detect whether a liver is useful for transplantation

Researchers from Vall d’Hebron Barcelona Hospital Campus and Pompeu Fabra University (UPF) have developed an artificial intelligence algorithm that allows knowing the levels of liver fat. The technology, based on photographs of the organ taken with a mobile phone, will help to know if a liver is suitable for transplantation in a more efficient way.

Every year, more than 1,000 liver transplants are performed in the State. This data makes the liver the second organ in the ranking of transplants, only after the kidney. Even so, in many cases, candidate livers have to be discarded because they are not suitable for transplantation. One of the causes is steatosis or fat accumulation in the liver. On these occasions, the cold that is necessary for the preservation of the organ causes the fat to form small crystals that break the cells. This can cause the liver to be severely damaged and therefore the transplant is not successful.

Therefore, it is necessary to analyze fat levels in the liver to be able to make a decision on whether the transplant can be performed or not. However, this assessment is usually done with the naked eye of a surgeon, only observing the tone of yellow, which is characteristic of fat. In this context, researchers from the Digestive and Liver Diseases area of VHIR, the Hepatobiolopancreatic Surgery and Transplants Service at Vall d’Hebron University Hospital and the BCN MedTech research Unit and the Department of Information and Communication Technologies at UPF, have developed Livercolor, a tool that allows to objectively assess the degree of steatosis from photographs of the liver.

Artificial intelligence for more efficient transplants

Take a photograph with the mobile phone of the liver’s transplantation candidate, use the artificial intelligence algorithm based on color and texture characteristics, and let the tool report fat levels and whether the organ is valid for transplantation or not. This is the idea of the creators of Livercolor.

The artificial tool on which the technology is based has been trained with photographs of the liver labeled according to the result of the liver biopsy, which is the most reliable way to know the degree of steatosis. “In the clinic, Livercolor would be able to determine whether a particular yellow tone and texture indicate a level of steatosis suitable for transplantation or not. In case it is not suitable, the liver is discarded”, explains Dr. Concepción Gómez-Gavara, surgeon of the Hepatobiolopancreatic Surgery and Transplantation Service at Vall d’Hebron University Hospital and researcher in the Digestive and Liver Diseases area at VHIR.

The use of Livercolor is, therefore, a more objective methodology when deciding whether it is necessary to discard the liver or not, compared to visual assessment. “This technology would allow to use a greater number of livers, which would otherwise be discarded. It is a clearly more efficient method than the decision based on what the doctor sees with the naked eye”, explains Dr. Itxarone Bilbao, head of Section of the Hepatobiolopancreatic Surgery and Transplantation Service at Vall d’Hebron University hospital. “Now we are working on a mobile app to be able to carry out a multicenter study that allows the technology to be applied to patients in the near future”, adds Dr. Gemma Piella, researcher at the BCN MedTech Research Unit at UPF. “We trust that once the project is developed it will end up being useful in a general way and will facilitate the process in liver donations”, concludes Javier Vázquez-Corral, researcher at the Department of Information and Communication Technologies at UPF.

Currently, Livercolor participates in the Barcelona Activa Pre-Acceleration Program, an initiative to support projects with high technological impact in their market entry process. Livercolor was one of the ten projects selected to participate in this edition of the program. In addition, the project has received €110,000 in funding from Fundación Mutua Madrileña through the XVII Health Research Grants, which will allow the Livercolor platform to continue developing.