Oral squamous cell carcinoma (OSCC) has the highest incidence among all the oral neoplasms, thus widely affects people’s life. Genetics has an important effect on the OSCC’s etiopathogenesis. However, OSCC’s molecular mechanism has not been completely investigated yet. We propose a novel method to detect genes that are associated to OSCC, where mean difference and variance difference are compared between cases and controls. This method can discover knowledge from gene data by processing multi-format information and can identify disease-associated genes for many complex human diseases, like melanoma, head and neck squamous cell carcinoma (HNSCC), lung adenocarcinoma, and serous ovarian cancer. Based on Gene Expression Omnibus, two OSCC datasets are obtained, and gene probes are then identified, typed as differentially variable (DV) and differentially expressed (DE), with numbers of 456 and 2375, respectively. Those probes can be further discriminated as DE-only, DV-only, and then DE-and-DV probes, with numbers of 2193, 274, and 182, respectively. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis is performed to separately analyze the genes that are relevant to these three kinds of specific probes, as per the DAVID functional analysis, implying they have diverse functions in OSCC. In the future, our novel method also can be applied to investigate other complex human diseases’ genetic risk factors.

UreteroPelvic Junction Obstruction (UPJO) is a common hydronephrosis disease in children that can result in an even progressive loss of renal function. Ultrasonography is an economical, radiationless, noninvasive, and high noise preliminary diagnostic step for UPJO. Artificial intelligence has been widely applied to medical fields and can greatly assist doctors’ diagnostic abilities. The demand for a highly secure network environment in transferring electronic medical data online, therefore, has led to the development of blockchain technology. In this study, we built and tested a framework that integrates a deep learning diagnosis model with blockchain technology. Our diagnosis model is a combination of an attention-based pyramid semantic segmentation network and a discrete wavelet transformation-processed residual classification network. We also compared the performance between benchmark models and our models. Our diagnosis model outperformed benchmarks on the segmentation task and classification task with MIoU = 87.93, MPA = 93.52, and accuracy = 91.77%. For the blockchain system, we applied the InterPlanetary File System protocol to build a secure and private sharing environment. This framework can automatically grade the severity of UPJO using ultrasound images, guarantee secure medical data sharing, assist in doctors’ diagnostic ability, relieve patients’ burden, and provide technical support for future federated learning and linkage of the Internet of Medical Things (IoMT).