Ultrasound-Based Nomogram for Predicting the Aggressiveness of Papillary Thyroid Carcinoma in Adolescents and Young Adults

Multimodal GPT model for assisting thyroid nodule diagnosis and management

Although using artificial intelligence (AI) to analyze ultrasound images is a promising approach to assessing thyroid nodule risks, traditional AI models lack transparency and interpretability. We developed a multimodal generative pre-trained transformer for thyroid nodules (ThyGPT), aiming to provide a transparent and interpretable AI copilot model for thyroid nodule risk assessment and management. Ultrasound data from 59,406 patients across nine hospitals were retrospectively collected to train and test the model. After training, ThyGPT was found to assist in reducing biopsy rates by more than 40% without increasing missed diagnoses. In addition, it detects errors in ultrasound reports 1,610 times faster than humans. With the assistance of ThyGPT, the area under the curve for radiologists in assessing thyroid nodule risks improved from 0.805 to 0.908 (p < 0.001). As an AI-generated content-enhanced computer-aided diagnosis (AIGC-CAD) model, ThyGPT has the potential to revolutionize how radiologists use such tools.

Accuracy of Radiomics in the Identification of Extrathyroidal Extension and BRAFV600E Mutations in Papillary Thyroid Carcinoma: A Systematic Review and Meta-analysis

RATIONALE AND OBJECTIVES

Extrathyroidal extension (ETE) and BRAFV600E mutation in papillary thyroid cancer (PTC) increase mortality and recurrence risk. Preoperative identification presents considerable challenges. Although radiomics has emerged as a potential tool for identifying ETE and BRAFV600E mutation, systematic evidence supporting its effectiveness remains insufficient. Therefore, this paper aims to determine the effectiveness of radiomics in detecting ETE and BRAFV600E mutations in PTC.

MATERIALS AND METHODS

PubMed, Web of Science, Cochrane, and Embase databases were searched until May 7th, 2024. The Radiomics Quality Score tool assessed bias risk. Subgroup analyses based on radiomics and clinical characteristics were conducted.

RESULTS

Our systematic review included 19 studies, encompassing 5337 PTC cases. Among these, 12 articles focused on ETE and seven articles focused on BRAFV600E mutations. For the identification of ETE in the validation set, the summarized machine learning (ML) models demonstrated 0.80c-index (95%CI: 0.77-0.83), 0.77 sensitivity (95%CI: 0.72-0.81), and 0.78 specificity (95%CI: 0.73-0.82). Radiomics based on ultrasound demonstrated 0.82c-index (95%CI: 0.78-0.86), 0.77 sensitivity (95%CI: 0.68-0.84), and 0.84 specificity (95%CI: 0.75-0.91). For the identification of BRAFV600E mutations in the validation set, the summarized ML models showed 0.80c-index (95%CI: 0.72-0.87), 0.76 sensitivity (95%CI: 0.67-0.84), and 0.88 specificity (95%CI: 0.77-0.94). ML models based on ultrasound-guided radiomics had 0.81c-index (95%CI: 0.74-0.89), 0.79 sensitivity (95%CI: 0.71-0.86), and 0.87 specificity (95%CI: 0.74-0.94).

CONCLUSION

Radiomics in identifying ETE and BRAFV600E mutation have high c-index, sensitivity, and specificity, especially images from ultrasound, demonstrating the potential for diagnosing ETE and BRAFV600E mutations in PTC.

Deep learning model for diagnosis of thyroid nodules with size less than 1 cm: A multicenter, retrospective study

Objective

To develop a ultrasound images based dual-channel deep learning model to achieve accurate early diagnosis of thyroid nodules less than 1 cm.

Methods

A dual-channel deep learning model called thyroid nodule transformer network (TNT-Net) was proposed. The model has two input channels for transverse and longitudinal ultrasound images of thyroid nodules, respectively. A total of 9649 nodules from 8455 patients across five hospitals were retrospectively collected. The data were divided into a training set (8453 nodules, 7369 patients), an internal test set (565 nodules, 512 patients), and an external test set (631 nodules, 574 patients).

Results

TNT-Net achieved an area under the curve (AUC) of 0.953 (95 % confidence interval (CI): 0.934, 0.969) on the internal test set and 0.941 (95 % CI: 0.921, 0.957) on the external test set, significantly outperforming traditional deep convolutional neural network models and single-channel swin transformer model, whose AUCs ranged from 0.800 (95 % CI: 0.759, 0.837) to 0.856 (95 % CI: 0.819, 0.881). Furthermore, feature heatmap visualization showed that TNT-Net could extract richer and more energetic malignant nodule patterns.

Conclusion

The proposed TNT-Net model significantly improved the recognition capability for thyroid nodules with size less than 1 cm. This model has the potential to reduce overdiagnosis and overtreatment of such nodules, providing essential support for precise management of thyroid nodules while complementing fine-needle aspiration biopsy.

Ultrasound-Base Radiomics for Discerning Lymph Node Metastasis in Thyroid Cancer: A Systematic Review and Meta-analysis

PURPOSE

Ultrasound is the imaging modality of choice for preoperative diagnosis of lymph node metastasis (LNM) in thyroid cancer (TC), yet its efficacy remains suboptimal. As radiomics gains traction in tumor diagnosis, its integration with ultrasound for LNM differentiation in TC has emerged, but its diagnostic merit is debated. This study assesses the accuracy of ultrasound-integrated radiomics in preoperatively diagnosing LNM in TC.

METHODS

Literatures were searched in PubMed, Embase, Cochrane, and Web of Science until July 11, 2023. Quality of the studies was assessed by the radiomics quality score (RQS). A meta-analysis was executed using a bivariate mixed effects model, with a subgroup analysis based on modeling variables (clinical features, radiomics features, or their combination).

RESULTS

Among 27 articles (16,410 TC patients, 6356 with LNM), the average RQS was 16.5 (SD:5.47). Sensitivity of the models based on clinical features, radiomics features, and radiomics features plus clinical features were 0.64, 0.76 and 0.69. Specificities were 0.77, 0.78 and 0.82. SROC values were 0.76, 0.84 and 0.81.

CONCLUSION

Ultrasound-based radiomics effectively evaluates LNM in TC preoperatively. Adding clinical features does not notably enhance the model's performance. Some radiomics studies showed high bias, possibly due to the absence of standard application guidelines.