ABSTRACT
Previously, the discrimination of collagen types I and II was successfully achieved using peptide pitch angle and anisotropic parameter methods. However, these methods require fitting polarization second harmonic generation (SHG) pixel-wise information into generic mathematical models, revealing inconsistencies in categorizing collagen type I and II blend hydrogels. In this study, a ResNet approach based on multipolarization SHG imaging is proposed for the categorization and regression of collagen type I and II blend hydrogels at 0%, 25%, 50%, 75%, and 100% type II, without the need for prior time-consuming model fitting. A ResNet model, pretrained on 18 progressive polarization SHG images at 10° intervals for each percentage, categorizes the five blended collagen hydrogels with a mean absolute error (MAE) of 0.021, while the model pretrained on nonpolarization images exhibited 0.083 MAE. Moreover, the pretrained models can also generally regress the blend hydrogels at 20%, 40%, 60%, and 80% type II. In conclusion, the multipolarization SHG image-based ResNet analysis demonstrates the potential for an automated approach using deep learning to extract valuable information from the collagen matrix.
Subject(s)
Collagen Type I , Hydrogels , Collagen , Diagnostic Imaging , Image Processing, Computer-AssistedABSTRACT
Purpose: Liver transplantation (LT) is the definite curative treatment for hepatocellular carcinoma (HCC), but recurrence can occur even under stringent criteria. "Delayed" HCC recurrence (>3 years after LT) is not common. Here, we present the clinical features of patients who developed delayed HCC recurrence after LT. Patients and Methods: We reviewed the data of eligible patients from February 1999 to December 2020 from medical records. Results: From among 195 (17%) HCC patients who received LT, 34 experienced HCC recurrence, with 5 (15%) delayed recurrence. These five explant tumors were staged T1b-T2, graded II-III, with two vascular invasion and four beyond the Milan criteria. The median time to recurrence was 6.1 years, with the longest interval being nearly 18 years. Recurrence patterns included two extrahepatic, one intrahepatic, and two mixed extrahepatic and intrahepatic recurrences. A drastic increase in serum alpha-fetoprotein levels was observed in four cases 1 year before recurrence. Management of recurrence included locoregional (surgssical resection in three, radiotherapy in three, transarterial chemoembolization in one, radiofrequency ablation in one) and systemic sorafenib use in three. Two patients died within 12-18 months, one died within 18-24 months, and two are still alive until the end of the study, with respective 13.5- and 16.5-month survival. Conclusion: Delayed HCC recurrence could occur over 10 years. Therefore, continual surveillance for recurrence is justified, but biomarkers and intervals or intensities specific for delayed recurrence are not validated, which warrants further validation to facilitate personalized medical care.