An Interpretable Multimodal Model Based on Cytology Ultrasound Noninvasively Predicts Lymph Node Metastasis in cN0 Papillary Thyroid Cancer
An Interpretable Multimodal Model Based on Cytology Ultrasound Noninvasively Predicts Lymph Node Metastasis in cN0 Papillary Thyroid Cancer
Rising thyroid carcinoma rates, with papillary thyroid carcinoma (PTC) as the main type (85-90% of cases), often show early cervical lymph node spread. This increases the risk of PTC patients for recurrence and death. A new study's multimodal model fuses preoperative US and cytology images to better predict lymph node metastasis, aiming to improve treatment plans, reduce unnecessary surgeries, and enhance patient outcomes.
Thyroid carcinoma incidence has been on the rise in recent years. Papillary thyroid carcinoma is the most prevalent type of differentiated thyroid carcinoma, accounting for 85% to 90% of malignant thyroid tumors. Despite its indolent nature, cervical lymph node metastasis is frequently observed at an early stage. Papillary thyroid carcinoma patients with cervical lymph node metastasis face an elevated risk of recurrence, distant metastasis, and mortality. Thus, non-invasive preoperative prediction of cervical lymph node metastasis is particularly vital for guiding treatment plans and prognostic. this study has developed a multimodal model integrating preoperative US images with cytological images of papillary thyroid carcinoma patients. The aim is to enhance the predictive accuracy for cervical lymph node metastasis, reduce unnecessary lymph node dissections, and provide real-time, precise guidance for determining the extent of surgical resection and prognostic assessment. This approach aims to optimize patient treatment strategies and enhance therapeutic outcomes.
Inclusion Criteria:
Exclusion Criteria: