Deep learning models for thyroid nodules diagnosis of fine-needle aspiration biopsy: a retrospective, prospective, multicentre study in China.

BACKGROUND: Accurately distinguishing between malignant and benign thyroid nodules through fine-needle aspiration cytopathology is crucial for appropriate therapeutic intervention. However, cytopathologic diagnosis is time consuming and hindered by the shortage of experienced cytopathologists. Reliable assistive tools could improve cytopathologic diagnosis efficiency and accuracy. We aimed to develop and test an artificial intelligence (AI)-assistive system for thyroid cytopathologic diagnosis according to the Thyroid Bethesda Reporting System. METHODS: 11 254 whole-slide images (WSIs) from 4037 patients were used to train deep learning models. Among the selected WSIs, cell level was manually annotated by cytopathologists according to The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) guidelines of the second edition (2017 version). A retrospective dataset of 5638 WSIs of 2914 patients from four medical centres was used for validation. 469 patients were recruited for the prospective study of the performance of AI models and their 537 thyroid nodule samples were used. Cohorts for training and validation were enrolled between Jan 1, 2016, and Aug 1, 2022, and the prospective dataset was recruited between Aug 1, 2022, and Jan 1, 2023. The performance of our AI models was estimated as the area under the receiver operating characteristic (AUROC), sensitivity, specificity, accuracy, positive predictive value, and negative predictive value. The primary outcomes were the prediction sensitivity and specificity of the model to assist cyto-diagnosis of thyroid nodules. FINDINGS: The AUROC of TBSRTC III+ (which distinguishes benign from TBSRTC classes III, IV, V, and VI) was 0·930 (95% CI 0·921-0·939) for Sun Yat-sen Memorial Hospital of Sun Yat-sen University (SYSMH) internal validation and 0·944 (0·929 - 0·959), 0·939 (0·924-0·955), 0·971 (0·938-1·000) for The First People's Hospital of Foshan (FPHF), Sichuan Cancer Hospital & Institute (SCHI), and The Third Affiliated Hospital of Guangzhou Medical University (TAHGMU) medical centres, respectively. The AUROC of TBSRTC V+ (which distinguishes benign from TBSRTC classes V and VI) was 0·990 (95% CI 0·986-0·995) for SYSMH internal validation and 0·988 (0·980-0·995), 0·965 (0·953-0·977), and 0·991 (0·972-1·000) for FPHF, SCHI, and TAHGMU medical centres, respectively. For the prospective study at SYSMH, the AUROC of TBSRTC III+ and TBSRTC V+ was 0·977 and 0·981, respectively. With the assistance of AI, the specificity of junior cytopathologists was boosted from 0·887 (95% CI 0·8440-0·922) to 0·993 (0·974-0·999) and the accuracy was improved from 0·877 (0·846-0·904) to 0·948 (0·926-0·965). 186 atypia of undetermined significance samples from 186 patients with BRAF mutation information were collected; 43 of them harbour the BRAFV600E mutation. 91% (39/43) of BRAFV600E-positive atypia of undetermined significance samples were identified as malignant by the AI models. INTERPRETATION: In this study, we developed an AI-assisted model named the Thyroid Patch-Oriented WSI Ensemble Recognition (ThyroPower) system, which facilitates rapid and robust cyto-diagnosis of thyroid nodules, potentially enhancing the diagnostic capabilities of cytopathologists. Moreover, it serves as a potential solution to mitigate the scarcity of cytopathologists. FUNDING: Guangdong Science and Technology Department. TRANSLATION: For the Chinese translation of the abstract see Supplementary Materials section.

Artificial intelligence in cancer pathology: Challenge to meet increasing demands of precision medicine.

Clinical efforts on precision medicine are driving the need for accurate diagnostic, new prognostic and novel drug predictive assays to inform patient selection and stratification for disease treatment. Accumulating evidence suggests that a combination of cancer pathology and artificial intelligence (AI) can meet this requirement. In the present review, the past, present and emerging integrations of AI into cancer pathology were comprehensively reviewed, which were divided into four main groups to highlight the roles of AI­integrated cancer pathology in precision medicine. Furthermore, the unsolved problems and future challenges in AI­integrated cancer pathology were also discussed. It was found that, although AI­integrated cancer pathology could enable the amalgamation of complex morphological phenotypes with the multi­omics datasets that drove precision medicine, synergies of cancer pathology with other medical tools could be more promising for the clinic when making an accurate and rapid decision in personalized treatments for patients. It was hypothesized by the authors that exploring the potential advantages of the multimodal integration of cancer pathology, imaging­omics, protein­omics and other­omics, as well as clinical data to decide upon appropriate management and improve patient outcomes may be the most challenging issue of cancer precision medicine in the future.

Exploring the Pharmacological Action and Molecular Mechanism of Salidroside in Inhibiting MCF-7 Cell Proliferation and Migration.

Salidroside (Sal) contains anti-carcinogenic, anti-hypoxic, and anti-inflammatory pharmacological activities. However, its underlying anti-breast cancer mechanisms have been only incompletely elucidated. Hence, this protocol intended to decode the potential of Sal in regulating the PI3K-AKT-HIF-1α-FoxO1 pathway in the malignant proliferation of human breast cancer MCF-7 cells. First, the pharmacological activity of Sal against MCF-7 was evaluated by CCK-8 and cell scratch assays. Moreover, the resistance of MCF-7 cells was measured by migration and Matrigel invasion assays. For cell apoptosis and cycle assays, MCF-7 cells were processed in steps with annexin V-FITC/PI and cell cycle-staining detection kits for flow cytometry analyses, respectively. The levels of reactive oxygen species (ROS) and Ca2+ were examined by DCFH-DA and Fluo-4 AM immunofluorescence staining. The activities of Na+-K+-ATPase and Ca2+-ATPase were determined using the corresponding commercial kits. The protein and gene expression levels in apoptosis and the PI3K-AKT-HIF-1α-FoxO1 pathway were further determined using western blot and qRT-PCR analyses, respectively. We found that Sal treatment significantly restricted the proliferation, migration, and invasion of MCF-7 cells with dose-dependent effects. Meanwhile, Sal administration also dramatically forced MCF-7 cells to undergo apoptosis and cell cycle arrest. The immunofluorescence tests showed that Sal observably stimulated ROS and Ca2+ production in MCF-7 cells. Further data confirmed that Sal promoted the expression levels of pro-apoptotic proteins, Bax, Bim, cleaved caspase-9/7/3, and their corresponding genes. Consistently, Sal intervention prominently reduced the expression of the Bcl-2, p-PI3K/PI3K, p-AKT/AKT, mTOR, HIF-1α, and FoxO1 proteins and their corresponding genes. In conclusion, Sal can be used as a potential herb-derived compound for treating breast cancer, as it may reduce the malignant proliferation, migration, and invasion of MCF-7 cells by inhibiting the PI3K-AKT-HIF-1α-FoxO1 pathway.