RÉSUMÉ
BACKGROUND: Postoperative hypoparathyroidism is a major complication of thyroidectomy, occurring when the parathyroid glands are inadvertently damaged during surgery. Although intraoperative images are rarely used to train artificial intelligence (AI) because of its complex nature, AI may be trained to intraoperatively detect parathyroid glands using various augmentation methods. The purpose of this study was to train an effective AI model to detect parathyroid glands during thyroidectomy. METHODS: Video clips of the parathyroid gland were collected during thyroid lobectomy procedures. Confirmed parathyroid images were used to train three types of datasets according to augmentation status: baseline, geometric transformation, and generative adversarial network-based image inpainting. The primary outcome was the average precision of the performance of AI in detecting parathyroid glands. RESULTS: 152 Fine-needle aspiration-confirmed parathyroid gland images were acquired from 150 patients who underwent unilateral lobectomy. The average precision of the AI model in detecting parathyroid glands based on baseline data was 77%. This performance was enhanced by applying both geometric transformation and image inpainting augmentation methods, with the geometric transformation data augmentation dataset showing a higher average precision (79%) than the image inpainting model (78.6%). When this model was subjected to external validation using a completely different thyroidectomy approach, the image inpainting method was more effective (46%) than both the geometric transformation (37%) and baseline (33%) methods. CONCLUSION: This AI model was found to be an effective and generalizable tool in the intraoperative identification of parathyroid glands during thyroidectomy, especially when aided by appropriate augmentation methods. Additional studies comparing model performance and surgeon identification, however, are needed to assess the true clinical relevance of this AI model.
RÉSUMÉ
Purpose: Intraoperative neurophysiological monitoring (IONM) has been introduced in thyroid surgery to prevent injury of the recurrent laryngeal nerve (RLN). However, its effectiveness remains controversial in robotic thyroidectomy (RT). This study aimed to compare the surgical outcome of RT in patients with and without the application of IONM. Methods: This retrospective case-control study included 100 patients who underwent total thyroidectomy via robotic bilateral axillo-breast approach in a tertiary center. A study group of 50 patients who had IONM during RT was compared to a control group of 50 patients who underwent RT with nerve visualization alone. Results: The sex ratio (4:45 vs. 7:43, P = 0.538), mean age (39.3 ± 7.1 years vs. 37.5 ± 10.4 years, P = 0.304), and body mass index (23.1 ± 2.6 kg/m2 vs. 22.2 ± 3.9 kg/m2, P = 0.215) were comparable between the IONM and control groups. Pathologic features including tumor size (0.8 cm vs. 0.9 cm, P = 0.283), extrathyroidal extension (58.0% vs. 24.0%, P = 0.316), lymph node metastasis (30% vs. 34%, P = 0.668), and number of lymph nodes (5.3 vs. 5.3, P = 0.668) showed no differences. There was no permanent RLN palsy, postoperative bleeding, and wound complications. Transient hypoparathyroidism was observed in 12 (24.0%) and 14 (28.0%), permanent hypoparathyroidism in 0 (0%) and 1 (2.0%), and transient RLN palsy was observed in 3 (6.0%) and 3 (6.0%), respectively. Conclusion: We did not demonstrate a clear advantage of IONM in RT. Controversies regarding the effectiveness of IONM is not closed.
RÉSUMÉ
BACKGROUND: There is a noticeable lack of information on the levels of both non-essential and essential trace elements in women aged over 50. The main objective of this study is to investigate trace element concentrations and explore the influence of sociodemographic factors and dietary sources of exposure in this demographic. METHODS: We analyzed 19 trace elements, including manganese, cobalt, copper, zinc, molybdenum, chromium, nickel, arsenic, strontium, cadmium, tin, antimony, cesium, barium, tungsten, mercury, thallium, lead, and uranium, using ICP-MS and mercury analyzer. Urine samples were obtained from a cohort of 851 women aged over 50 who participated in the 8th KoGES-Ansung study (2017-2018). Multiple linear models were employed to explore associations between urinary trace element concentrations and sociodemographic factors and dietary sources of exposure. We used K-means clustering to discern patterns of exposure to trace elements and identify contributing factors and sources. RESULTS: Our findings indicate higher concentrations of molybdenum (Mo), arsenic (As), cadmium (Cd), and lead (Pb) in our study population compared to women in previous studies. The study population were clustered into two distinct groups, characterized by lower or higher urinary concentrations. Significant correlations between age and urinary concentrations were observed in Ni. Smoking exhibited positive associations with urinary Cd and As. Associations with dietary sources of trace elements were more distinct in women in the high-exposure group. Urinary antimony (Sb) was positively linked to mushroom and egg intake, As to mushroom and fish, and Hg to egg, dairy products, fish, seaweed, and shellfish. CONCLUSIONS: Our study underscores the significant gap in understanding urinary concentrations of trace elements in women aged over 50. With higher concentrations of certain elements compared to previous studies and significant correlations between age, smoking, and specific food sources, it is imperative to address this gap through targeted dietary source-specific risk management.