J Formos Med Assoc
-
The accuracy of histopathology diagnosis largely depends on the pathologist's experience. It usually takes over 10 years to cultivate a senior pathologist, and small numbers of them lead to a high workload for those available. Meanwhile, inconsistent diagnostic results may arise among different pathologists, especially in complex cases, because diagnosis based on morphology is subjective. Computerized analysis based on deep learning has shown potential benefits as a diagnostic strategy. ⋯ For AI to assist pathologists in daily practice, to help a pathologist making a definite diagnosis is not the prime purpose at present time. The benefits could come from cancer screening and double-check quality control in a heavy workload which could distract the attention of pathologist during the time constraint examination process. We propose a two-steps method to identify cancerous areas in endoscopic gastric biopsy slices via deep learning. Then a 3D model is used to further mark all the positions of GC in the picture, and the model overcomes the problem that deep learning can't catch all GC.
-
Alanine aminotransferase (ALT) is a cost-effective screening test for asymptomatic liver diseases. The aims of this study are to redefine the ULNs of ALT using the 2010-2012 Nutrition and Health Survey in Taiwan (NAHSIT) database and to determine whether the updated ULNs can better screen for metabolic dysfunction-associated fatty liver disease (MAFLD) in obese children. ⋯ After taking into account MAFLD-related metabolic risk factors, the ULNs of ALT are 23 IU/L for boys and 18 IU/L for girls in Taiwan. The updated ULNs may be better cutoffs for screening MAFLD in obese children.