上海长海医院

The classification of pancreatic cystic neoplasms into benign and malignant categories is crucial for medical decision-making.This paper is dedicated to enhancing the accuracy of pancreatic cystic neoplasms classification to assist physicians in formulating more precise diagnostic and therapeutic plans.Utilizing radiomics technology and the ResNet50 neural network,a novel classification method for pancreatic cystic neoplasms is proposed,integrating multi-kernel learning and multi-source feature fusion.The key steps of this method include feature selection,kernel matrix fusion,and the construction of the classification model.Feature selection is performed using the least absolute shrinkage and selection operator(LASSO)to reduce redundant features and improve the model's generalization ability.Subsequently,multi-source features,screened through feature selection,are mapped in basic kernel functions to construct basic kernel matrices for multi-source features.The weights of these kernel matrices are then optimized and summed up to form a fused kernel matrix.Finally,a support vector machine(SVM)classifier is utilized to categorize pancreatic serous and mucinous cystic tumors.The significance of this process lies in SVM's ability to use the kernel matrix for inner product operations in high-dimensional spaces,thereby finding a hyperplane to classify data in such spaces.The fused kernel matrix,containing multi-source information after feature mapping,provides higher-dimensional and more complex feature representations.Experimental results demonstrate significant performance improvements in the classification task of pancreatic cystic neoplasms,offering more reliable auxiliary information to physicians and holding substantial clinical application potential.

[Objectives] Pedigree investigation and predictive genetic testing for colonic adenomatous polyposis (CAP) has not been systematically conducted in China, and its practical value remains unreported. This study aims to explore the practical significance of predictive genetic testing for families affected by CAP in China. [Methods] A cross-sectional study, conducted by the Screening and Prevention Center for Hereditary Colorectal Cancer at Changhai Hospital, aimed to perform predictive genetic testing for families affected by APC-associated polyposis (AAP) through pedigree investigation. Clinical information and pedigree investigation of the probands focused on core details such as the clinical classification of colon polyposis. Subjects for predictive genetic testing included first-degree relatives who could be traced. The testing methods for APC gene encompassed Sanger sequencing, second-generation sequencing, and multiplex ligation-dependent probe amplification. Detected APC gene mutations comprised oligonucleotide variations and large fragment deletions. The optimal timing for conducting predictive genetic testing was explored considering the detection rate of gene mutations stratified by age. [Results] The study included a total of 65 families affected by AAP. The probands exhibited diverse clinical manifestations, including colon adenoma, gastric and duodenal adenoma, thyroid cancer, congenital hypertrophy of the retinal pigment epithelium, abnormalities of jaw and tooth, and hard fibroma. Predictive genetic testing was conducted on 130 asymptomatic relatives (61 males). Among the 65 families, a total of 49 germline mutations in the APC gene were detected, including 13 mutations that had not been previously reported. The median age of the subjects was 12 (range: 6 to 24 years old), with 52 mutation carriers detected among family members, resulting in a mutation detection rate of 40%. When restricting the age of the subjects to the overall age limit of 50 years (in 5-year-increments), the mutation detection rates were 53.8%, 48.2%, 52.1%, 50.6%, 47.6%, 44.2%, 43.0%, 41.6%, 40.9%, and 40.3%, respectively. Mutation detection rates occurred at notable ages 7, 14, and 18 years with rates of 52.6%, 52.1%, and 48.8%, respectively. In the descendant group, the mutation detection rate was 44.1% (52/118). [Conclusion] For families affected by CAP with clear pathogenic mutations likes AAP, predictive genetic testing proves valuable in identifying carriers of these mutations early, facilitating prompt diagnosis and treatment. The mutation detection rate in predictive genetic testing typically demonstrates an increasing trend with decreasing age of the subjects. Specifically, when the age of the subjects is less than or equal to 25 years old, the detection rate approaches 50%.