Machine learning on MRI radiomic features: identification of molecular subtype alteration in breast cancer after neoadjuvant therapy.

OBJECTIVES: Recent studies have revealed the change of molecular subtypes in breast cancer (BC) after neoadjuvant therapy (NAT). This study aims to construct a non-invasive model for predicting molecular subtype alteration in breast cancer after NAT. METHODS: Eighty-two estrogen receptor (ER)-negative/ human epidermal growth factor receptor 2 (HER2)-negative or ER-low-positive/HER2-negative breast cancer patients who underwent NAT and completed baseline MRI were retrospectively recruited between July 2010 and November 2020. Subtype alteration was observed in 21 cases after NAT. A 2D-DenseUNet machine-learning model was built to perform automatic segmentation of breast cancer. 851 radiomic features were extracted from each MRI sequence (T2-weighted imaging, ADC, DCE, and contrast-enhanced T1-weighted imaging), both in the manual and auto-segmentation masks. All samples were divided into a training set (n = 66) and a test set (n = 16). XGBoost model with 5-fold cross-validation was performed to predict molecular subtype alterations in breast cancer patients after NAT. The predictive ability of these models was subsequently evaluated by the AUC of the ROC curve, sensitivity, and specificity. RESULTS: A model consisting of three radiomics features from the manual segmentation of multi-sequence MRI achieved favorable predictive efficacy in identifying molecular subtype alteration in BC after NAT (cross-validation set: AUC = 0.908, independent test set: AUC = 0.864); whereas an automatic segmentation approach of BC lesions on the DCE sequence produced good segmentation results (Dice similarity coefficient = 0.720). CONCLUSIONS: A machine learning model based on baseline MRI is proven useful for predicting molecular subtype alterations in breast cancer after NAT. KEY POINTS: • Machine learning models using MRI-based radiomics signature have the ability to predict molecular subtype alterations in breast cancer after neoadjuvant therapy, which subsequently affect treatment protocols. • The application of deep learning in the automatic segmentation of breast cancer lesions from MRI images shows the potential to replace manual segmentation..

Optimizing robotic thyroid surgery: lessons learned from an retrospective analysis of 104 cases.

Background: Robotic assistance in thyroidectomy is a developing field that promises enhanced surgical precision and improved patient outcomes. This study investigates the impact of the da Vinci Surgical System on operative efficiency, learning curve, and postoperative outcomes in thyroid surgery. Methods: We conducted a retrospective cohort study of 104 patients who underwent robotic thyroidectomy between March 2018 and January 2022. We evaluated the learning curve using the Cumulative Sum (CUSUM) analysis and analyzed operative times, complication rates, and postoperative recovery metrics. Results: The cohort had a mean age of 36 years, predominantly female (68.3%). The average body mass index (BMI) was within the normal range. A significant reduction in operative times was observed as the series progressed, with no permanent hypoparathyroidism or recurrent laryngeal nerve injuries reported. The learning curve plateaued after the 37th case. Postoperative recovery was consistent, with no significant difference in hospital stay duration. Complications were minimal, with a noted decrease in transient vocal cord palsy as experience with the robotic system increased. Conclusion: Robotic thyroidectomy using the da Vinci system has demonstrated a significant improvement in operative efficiency without compromising safety. The learning curve is steep but manageable, and once overcome, it leads to improved surgical outcomes and high patient satisfaction. Further research with larger datasets and longer follow-up is necessary to establish the long-term benefits of robotic thyroidectomy.

Intraoperative AI-assisted early prediction of parathyroid and ischemia alert in endoscopic thyroid surgery.

BACKGROUND: The preservation of parathyroid glands is crucial in endoscopic thyroid surgery to prevent hypocalcemia and related complications. However, current methods for identifying and protecting these glands have limitations. We propose a novel technique that has the potential to improve the safety and efficacy of endoscopic thyroid surgery. PURPOSE: Our study aims to develop a deep learning model called PTAIR 2.0 (Parathyroid gland Artificial Intelligence Recognition) to enhance parathyroid gland recognition during endoscopic thyroidectomy. We compare its performance against traditional surgeon-based identification methods. MATERIALS AND METHODS: Parathyroid tissues were annotated in 32 428 images extracted from 838 endoscopic thyroidectomy videos, forming the internal training cohort. An external validation cohort comprised 54 full-length videos. Six candidate algorithms were evaluated to select the optimal one. We assessed the model's performance in terms of initial recognition time, identification duration, and recognition rate and compared it with the performance of surgeons. RESULTS: Utilizing the YOLOX algorithm, we developed PTAIR 2.0, which demonstrated superior performance with an AP50 score of 92.1%. The YOLOX algorithm achieved a frame rate of 25.14 Hz, meeting real-time requirements. In the internal training cohort, PTAIR 2.0 achieved AP50 values of 94.1%, 98.9%, and 92.1% for parathyroid gland early prediction, identification, and ischemia alert, respectively. Additionally, in the external validation cohort, PTAIR outperformed both junior and senior surgeons in identifying and tracking parathyroid glands (p < 0.001). CONCLUSION: The AI-driven PTAIR 2.0 model significantly outperforms both senior and junior surgeons in parathyroid gland identification and ischemia alert during endoscopic thyroid surgery, offering potential for enhanced surgical precision and patient outcomes.

Accurate preoperative prediction of nodal metastasis in papillary thyroid microcarcinoma: Towards optimal management of patients.

OBJECTIVE: To enhance the accuracy in predicting lymph node metastasis (LNM) preoperatively in patients with papillary thyroid microcarcinoma (PTMC), refining the "low-risk" classification for tailored treatment strategies. METHODS: This study involves the development and validation of a predictive model using a cohort of 1004 patients with PTMC undergoing thyroidectomy along with central neck dissection. The data was divided into a training cohort (n = 702) and a validation cohort (n = 302). Multivariate logistic regression identified independent LNM predictors in PTMC, leading to the construction of a predictive nomogram model. The model's performance was assessed through ROC analysis, calibration curve analysis, and decision curve analysis. RESULTS: Identified LNM predictors in PTMC included age, tumor maximum diameter, nodule-capsule distance, capsular contact length, bilateral suspicious lesions, absence of the lymphatic hilum, microcalcification, and sex. Especially, tumors larger than 7 mm, nodules closer to the capsule (less than 3 mm), and longer capsular contact lengths (more than 1 mm) showed higher LNM rates. The model exhibited AUCs of 0.733 and 0.771 in the training and validation cohorts respectively, alongside superior calibration and clinical utility. CONCLUSION: This study proposes and substantiates a preoperative predictive model for LNM in patients with PTMC, honing the precision of "low-risk" categorization. This model furnishes clinicians with an invaluable tool for individualized treatment approach, ensuring better management of patients who might be proposed observation or ablative options in the absence of such predictive information.

CA10 is associated with HBV-related hepatocarcinogenesis.

Hepatocellular carcinoma (HCC) is the main threat for the patients infected with hepatitis B virus (HBV), but the oncogenic mechanism of HBV-related HCC is still controversial. Previously, we have found that several HBV surface gene (HBS) non-sense mutations are oncogenic. Among these mutations, sW182* was found to have the most potent oncogenicity. In this study, we found that Carbonic Anhydrase X (CA10) level was specifically increased in sW182* mutant-expressing cells. CA10 overexpression was also associated with HBS nonsense mutation in HBV-related HCC tumor tissues. Transformation and tumorigenesis assays revealed that CA10 had significant oncogenic activity. In addition, CA10 overexpression resulted in dysregulation of apoptosis-related proteins, including Mcl-1, Bcl-2, Bcl-xL and Bad. While searching for the regulatory mechanism of CA10, miR-27b was found to downregulate CA10 expression by regulating its mRNA degradation and its expression was decreased in sW182* mutant cells. Moreover, CA10 overexpression was associated with down-regulation of miR-27b in human HBV-related HCC tumor tissues with sW182* mutation. Therefore, induction of the expression of CA10 through repression of miR-27b by sW182* might be one mechanism involved in HBS mutation-related hepatocarcinogenesis.

Development of Artificial Intelligence for Parathyroid Recognition During Endoscopic Thyroid Surgery.

OBJECTIVE: We aimed to establish an artificial intelligence (AI) model to identify parathyroid glands during endoscopic approaches and compare it with senior and junior surgeons' visual estimation. METHODS: A total of 1,700 images of parathyroid glands from 166 endoscopic thyroidectomy videos were labeled. Data from 20 additional full-length videos were used as an independent external cohort. The YOLO V3, Faster R-CNN, and Cascade algorithms were used for deep learning, and the optimal algorithm was selected for independent external cohort analysis. Finally, the identification rate, initial recognition time, and tracking periods of PTAIR (Artificial Intelligence model for Parathyroid gland Recognition), junior surgeons, and senior surgeons were compared. RESULTS: The Faster R-CNN algorithm showed the best balance after optimizing the hyperparameters of each algorithm and was updated as PTAIR. The precision, recall rate, and F1 score of the PTAIR were 88.7%, 92.3%, and 90.5%, respectively. In the independent external cohort, the parathyroid identification rates of PTAIR, senior surgeons, and junior surgeons were 96.9%, 87.5%, and 71.9%, respectively. In addition, PTAIR recognized parathyroid glands 3.83 s ahead of the senior surgeons (p = 0.008), with a tracking period 62.82 s longer than the senior surgeons (p = 0.006). CONCLUSIONS: PTAIR can achieve earlier identification and full-time tracing under a particular training strategy. The identification rate of PTAIR is higher than that of junior surgeons and similar to that of senior surgeons. Such systems may have utility in improving surgical outcomes and also in accelerating the education of junior surgeons. LEVEL OF EVIDENCE: 3 Laryngoscope, 132:2516-2523, 2022.