Thyroid nodules are a common endocrine disorder, and their early detection and accurate diagnosis are crucial for the prevention of thyroid cancer. However, the highly heterogeneous morphology and boundaries of thyroid nodules pose significant challenges to their precise identification and classification. Traditional diagnostic approaches rely heavily on physicians' experience, which increases the risk of misdiagnosis and missed diagnoses. With the rapid advancement of computer-aided diagnosis (CAD) technologies, applying deep learning algorithms to the analysis of thyroid nodule ultrasound images has shown great potential. This paper reviews the latest research progress on deep learning-based CAD methods for thyroid nodules, with a focus on their applications in image preprocessing, segmentation and classification. The advantages and limitations of current techniques are analyzed, and potential future directions are discussed. This review aims to highlight the potential of deep learning in thyroid nodule diagnosis and to provide a foundation for selecting feasible pathways for future clinical applications.
Humans ; Thyroid Nodule/diagnostic imaging* ; Deep Learning ; Ultrasonography/methods* ; Diagnosis, Computer-Assisted/methods* ; Algorithms ; Thyroid Neoplasms/diagnostic imaging* ; Image Processing, Computer-Assisted/methods*

Oral squamous cell carcinoma (OSCC) accounts for over 90% of oral malignancies, with more than 370 000 new cases and approximately 188 000 deaths annually worldwide. In China, there are roughly 65 000 new cases and 35 000 deaths each year, showing a significant upward trend compared with 2015 statistics. Despite continuous advancements in treatment modalities, the 5-year survival rate remains stagnant at 50%-60%, where tumor heterogeneity and therapy resistance persist as fundamental barriers to precision oncology. To address these critical challenges, this study established a standardized bioban-king protocol for OSCC patient-derived organoids (PDOs) (Patent: Method for constructing an oral squamous cell carcinoma organoid bank, ZL202311378598.3). Through groundbreaking optimization of culture media, enzymatic digestion kinetics, and stepwise cryopreservation, we achieved a biobanking success rate exceeding 95% and pioneered synchronous cultivation of matched primary tumors, lymph node metastases, and adjacent normal mucosa from individual patients, preserving spatial heterogeneity and stromal interactions. Leveraging this platform, we developed high-throughput drug screening: Quantified heterogeneity-driven differential chemoresponse using adenosine triphosphate (ATP)-based viability assays; We discovered resistance mechanisms: Identified sialylated cancer IgG (SIA-cIgG)-mediated cis-platin resistance (primary/secondary) through PTPN13 suppression, with anti-SIA-cIgG combination therapy demonstrating synergistic efficacy. Besides, we elucidated metastatic drivers: CRISPR-Cas9-edited organoids revealed WDR54 promoted metastasis via H3K4me3/H4K16ac epigenetic reprogramming, activating epithelial-mesenchymal plasticity (EMP) and inducing partial epithelial-mesenchymal transition (pEMT). This "holographic patient-mirroring" platform provided unprecedented resolution for OSCC precision therapy and had been formally incorporated into the Chinese Stomatological Association Technical Guidelines (Technical guideline for establishing patient-derived oral squamous cell carcinoma organoid banks, CHSA 2024-08). Future integration of immune-competent organoids, 3D-bioprinted vasculature, and multi-omics-AI systems will accelerate personalized oncology. These innovations will accelerate clinical translation of personalized therapeutic regimens, ultimately bridging the gap between bench research and bedside application.
Humans ; Organoids/pathology* ; Mouth Neoplasms/genetics* ; Carcinoma, Squamous Cell/pathology* ; Tissue Banks ; Biological Specimen Banks

Objective To analyze the expression levels and clinical significance of N-myc down-stream-regulated gene 3(NDRG3)and semaphoring 3A(SEMA3A)in elderly patients with acute ischemic stroke(AIS).Methods A total of 100 elderly AIS patients admitted to Department of Geriatrics of Liaocheng People's Hospital from September 2020 to September 2022 were included as the study group.According to their NIHSS score at admission,they were divided into mild(34 cases),moderate(31 cases)and severe(35 cases)subgroups.All patients were followed up for 3 months after discharge.And they were assigned into good prognosis group(69 cases)and poor prognosis group(31 cases)based on the modified Rankin scale score.Another 100 healthy individ-uals who underwent physical examination in our hospital during the same period were recruited as the control group.Western blotting was used to detect the protein expression of NDRG3 and SEMA3A in peripheral blood mononuclear cells(PBMC).ELISA was applied to measure the con-tents of VEGF,TGF-1,TNF-α,and IL-17 in peripheral blood samples.Spearman rank correlation analysis was performed to analyze the correlation of NDRG3 and SEMA3A levels with NIHSS score,and ROC curve was plotted to analyze the values of NDRG3 and SEMA3A in predicting poor prognosis in elderly AIS patients.Results The expression levels of NDRG3 and SEMA3A in PBMC were obviously higher in the study group than the control group(1.11±0.16 vs 0.76± 0.13,0.78±0.13 vs 0.42±0.09,P＜0.01).The levels in the mild,moderate and severe subgroups were significantly higher than that of the control group(P＜0.01).The poor prognosis group had statistically higher expression levels of NDRG3 and SEMA3A than the good prognosis group(P＜0.01).Spearman rank correlation analysis showed that the NIHSS score was positively corre-lated with the expression levels of NDRG3 and SEMA3A in elderly AIS patients(r=0.597,P＜0.01;r=0.618,P＜0.01),while the NDRG3 level was positively correlated with that of SEMA3 A(r=0.477,P＜0.01).ROC curve analysis indicated that the AUC value of combined NDRG3 and SEMA3A levels was superior to that of NDRG3 and SEMA3A alone in predicting of poor progno-sis(0.962 vs 0.861,0.880,P＜0.01).Conclusion The levels of NDRG3 and SEMA3A proteins are up-regulated in elderly AIS patients,and are closely associated with the severity and prognosis of the disease.