Construction and validation of nomograms to reduce completion thyroidectomy by predicting lymph node metastasis in low-risk papillary thyroid carcinoma

Using the TSA-LSTM two-stage model to predict cancer incidence and mortality

Cancer, the second-leading cause of mortality, kills 16% of people worldwide. Unhealthy lifestyles, smoking, alcohol abuse, obesity, and a lack of exercise have been linked to cancer incidence and mortality. However, it is hard. Cancer and lifestyle correlation analysis and cancer incidence and mortality prediction in the next several years are used to guide people's healthy lives and target medical financial resources. Two key research areas of this paper are Data preprocessing and sample expansion design Using experimental analysis and comparison, this study chooses the best cubic spline interpolation technology on the original data from 32 entry points to 420 entry points and converts annual data into monthly data to solve the problem of insufficient correlation analysis and prediction. Factor analysis is possible because data sources indicate changing factors. TSA-LSTM Two-stage attention design a popular tool with advanced visualization functions, Tableau, simplifies this paper's study. Tableau's testing findings indicate it cannot analyze and predict this paper's time series data. LSTM is utilized by the TSA-LSTM optimization model. By commencing with input feature attention, this model attention technique guarantees that the model encoder converges to a subset of input sequence features during the prediction of output sequence features. As a result, the model's natural learning trend and prediction quality are enhanced. The second step, time performance attention, maintains We can choose network features and improve forecasts based on real-time performance. Validating the data source with factor correlation analysis and trend prediction using the TSA-LSTM model Most cancers have overlapping risk factors, and excessive drinking, lack of exercise, and obesity can cause breast, colorectal, and colon cancer. A poor lifestyle directly promotes lung, laryngeal, and oral cancers, according to visual tests. Cancer incidence is expected to climb 18-21% between 2020 and 2025, according to 2021. Long-term projection accuracy is 98.96 percent, and smoking and obesity may be the main cancer causes.

How Many Lymph Nodes are Enough in Thyroidectomy? A Cohort Study Based on Real-World Data

BACKGROUND

Thyroidectomy with only limited examination of lymph nodes is considered to pose potential risk for harboring occult nodal disease in patients with papillary thyroid cancer (PTC). However, the optimal number of examined lymph nodes (ELNs) in patients with PTC with clinically lateral lymph node metastasis (cN1b) remains unclear.

PATIENTS AND METHODS

Patients with cN1b PTC who underwent therapeutic neck dissection were retrospectively enrolled. A β-binomial distribution was utilized to calculate the likelihood of occult nodal disease as a function of total number of ELNs, and recurrence-free survival analysis was performed using the Kaplan-Meier method.

RESULTS

Together 982 patients met the inclusion criteria for this study, of which 853 patients had node-positive disease. The median ELN count was 23 (interquartile range 14-33). Increased ELN counts were associated with a decreased rate of occult nodal disease. The prevalence of nodal metastasis was 84%, while the corrected prevalence was 90%. The estimated probability of false-negative nodal disease was 67% for patients with PTC when only a single node was examined. Survival analysis revealed that populations with higher probability of occult nodal diseases experienced significantly higher recurrence rate. For patient with cN1b PTC, 20 ELNs were required to achieve 95% confidence of having no occult nodal disease. Minimum thresholds of 24, 14, 14, and 15 ELNs were selected for patients with pT1, pT2, pT3, and pT4 diseases, respectively.

CONCLUSIONS

Our findings robustly conclude that a minimum of 20 ELNs is essential to assess the quality of neck dissection and acquire accurate staging for patients with cN1b PTC.

Elevated Levels of Interleukin-18 are Associated with Lymph Node Metastasis in Papillary Thyroid Carcinoma

Interleukin-18 (IL-18) is a proinflammatory cytokine that primarily stimulates the Th1 immune response. IL-18 exhibits anticancer activity and has been evaluated in clinical trials as a potential cancer treatment. However, evidence suggests that it may also facilitate the development and progression of some cancers. So far, the impact of IL-18 on papillary thyroid cancer (PTC) has not been investigated. In this study, we found that the expression of IL-18 was significantly increased in PTC compared to normal thyroid tissue. Elevated IL-18 expression was closely associated with lymphovascular invasion and lymph node metastases. Furthermore, compared to PTC patients with no nodal metastasis, serum IL-18 levels were slightly increased in patients with 1-4 nodal metastases and significantly elevated in patients with 5 or more nodal metastases. The pro-metastatic effect of IL-18 may be attributed to the simultaneous increase in the expression of S100A10, a known factor that is linked to nodal metastasis in PTC. In addition, the activation of several pathways, such as the intestinal immune network for lgA production and Staphylococcus aureus infection, may be involved in the metastasis process. Taken together, IL-18 may trigger pro-metastatic activity in PTC. Therefore, suppressing the function of IL-18 rather than enhancing it appears to be a reasonable strategy for treating aggressive PTC.

Development and Validation of Clinical-Radiomics Nomogram for Preoperative Prediction of Central Lymph Node Metastasis in Papillary Thyroid Carcinoma

BACKGROUND

This investigation sought to create and verify a nomogram utilizing ultrasound radiomics and crucial clinical features to preoperatively identify central lymph node metastasis (CLNM) in patients diagnosed with papillary thyroid carcinoma (PTC).

METHODS

We enrolled 1069 patients with PTC between January 2022 and January 2023. All patients were randomly divided into a training cohort (n = 748) and a validation cohort (n = 321). We extracted 129 radiomics features from the original gray-scale ultrasound image. Then minimum Redundancy-Maximum Relevance and Least Absolute Shrinkage and Selection Operator regression were used to select the CLNM-related features and calculate the radiomic signature. Incorporating the radiomic signature and clinical risk factors, a clinical-radiomics nomogram was constructed using multivariable logistic regression. The predictive performance of clinical-radiomics nomogram was evaluated by calibration, discrimination, and clinical utility in the training and validation cohorts.

RESULTS

The clinical-radiomics nomogram which consisted of five predictors (age, tumor size, margin, lateral lymph node metastasis, and radiomics signature), showed good calibration and discrimination in both the training (AUC 0.960; 95% CI, 0.947-0.972) and the validation (AUC 0.925; 95% CI, 0.895-0.955) cohorts. Discrimination of the clinical-radiomics nomogram showed better discriminative ability than the clinical signature, radiomics signature, and conventional ultrasound model in both the training and validation cohorts. Decision curve analysis showed satisfactory clinical utility of the nomogram.

CONCLUSION

The clinical-radiomics nomogram incorporating radiomic signature and key clinical features was efficacious in predicting CLNM in PTC patients.

Cervical lymph node metastasis prediction from papillary thyroid carcinoma US videos: a prospective multicenter study

BACKGROUND

Prediction of lymph node metastasis (LNM) is critical for individualized management of papillary thyroid carcinoma (PTC) patients to avoid unnecessary overtreatment as well as undesired under-treatment. Artificial intelligence (AI) trained by thyroid ultrasound (US) may improve prediction performance.

METHODS

From September 2017 to December 2018, patients with suspicious PTC from the first medical center of the Chinese PLA general hospital were retrospectively enrolled to pre-train the multi-scale, multi-frame, and dual-direction deep learning (MMD-DL) model. From January 2019 to July 2021, PTC patients from four different centers were prospectively enrolled to fine-tune and independently validate MMD-DL. Its diagnostic performance and auxiliary effect on radiologists were analyzed in terms of receiver operating characteristic (ROC) curves, areas under the ROC curve (AUC), accuracy, sensitivity, and specificity.

RESULTS

In total, 488 PTC patients were enrolled in the pre-training cohort, and 218 PTC patients were included for model fine-tuning (n = 109), internal test (n = 39), and external validation (n = 70). Diagnostic performances of MMD-DL achieved AUCs of 0.85 (95% CI: 0.73, 0.97) and 0.81 (95% CI: 0.73, 0.89) in the test and validation cohorts, respectively, and US radiologists significantly improved their average diagnostic accuracy (57% vs. 60%, P = 0.001) and sensitivity (62% vs. 65%, P < 0.001) by using the AI model for assistance.

CONCLUSIONS

The AI model using US videos can provide accurate and reproducible prediction of cervical lymph node metastasis in papillary thyroid carcinoma patients preoperatively, and it can be used as an effective assisting tool to improve diagnostic performance of US radiologists.

TRIAL REGISTRATION

We registered on the Chinese Clinical Trial Registry website with the number ChiCTR1900025592.