AI Article Synopsis
- Several studies have explored the use of artificial intelligence (AI) in analyzing cytology images, but its implementation in clinical practice remains limited.
- The study assessed the effectiveness of AI-based image analysis for thyroid fine-needle aspiration cytology (FNAC) using a large dataset of over 148,000 images and the EfficientNetV2-L model.
- Results showed high accuracy for most thyroid conditions, with a precision-recall area under the curve exceeding 0.95 for many nodules, suggesting that AI could enhance clinical management, particularly for atypia of undetermined significance (AUS) and follicular neoplasms.
Article Abstract
Background: Several studies have used artificial intelligence (AI) to analyze cytology images, but AI has yet to be adopted in clinical practice. The objective of this study was to demonstrate the accuracy of AI-based image analysis for thyroid fine-needle aspiration cytology (FNAC) and to propose its application in clinical practice.
Methods: In total, 148,395 microscopic images of FNAC were obtained from 393 thyroid nodules to train and validate the data, and EfficientNetV2-L was used as the image-classification model. The 35 nodules that were classified as atypia of undetermined significance (AUS) were predicted using AI training.
Results: The precision-recall area under the curve (PR AUC) was >0.95, except for poorly differentiated thyroid carcinoma (PR AUC = 0.49) and medullary thyroid carcinoma (PR AUC = 0.91). Poorly differentiated thyroid carcinoma had the lowest recall (35.4%) and was difficult to distinguish from papillary thyroid carcinoma, medullary thyroid carcinoma, and follicular thyroid carcinoma. Follicular adenomas and follicular thyroid carcinomas were distinguished from each other by 86.7% and 93.9% recall, respectively. For two-dimensional mapping of the data using t-distributed stochastic neighbor embedding, the lymphomas, follicular adenomas, and anaplastic thyroid carcinomas were divided into three, two, and two groups, respectively. Analysis of the AUS nodules showed 94.7% sensitivity, 14.4% specificity, 56.3% positive predictive value, and 66.7% negative predictive value.
Conclusions: The authors developed an AI-based approach to analyze thyroid FNAC cases encountered in routine practice. This analysis could be useful for the clinical management of AUS and follicular neoplasm nodules (e.g., an online AI platform for thyroid cytology consultations).
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1002/cncy.22669 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Development and validation of a metabolic syndrome and its components to predict the efficacy of neoadjuvant chemotherapy in breast cancer: An observational, single-center, cohort study.
Medicine (Baltimore)
January 2025
Department of Thyroid and Breast Surgery, Hebei General Hospital, Shijiazhuang, Hebei Province, China.
To assess whether metabolic syndrome can be used as a reference index to evaluate the efficacy of neoadjuvant chemotherapy treatment for breast cancer (BC). Seventy cases of female BC patients who received neoadjuvant chemotherapy treatment and surgical treatment at the Glandular Surgery Department of Hebei Provincial People's Hospital from January 2021 to December 2023 were retrospectively collected, and clinical data such as puncture pathology were recorded. The clinical data were analyzed by 1-way analysis using the χ2 test, and further multifactorial logistic regression analysis was performed for statistically significant differences.
View Article and Find Full Text PDFEvaluating the Relationship between Cathepsins and Papillary Thyroid Carcinoma: A Mendelian Randomization Study.
Endocr Metab Immune Disord Drug Targets
January 2025
Department of Vascular and Thyroid Surgery, Affiliated Hospital of Guangdong Medical University, Guangdong, China.
Background: Papillary Thyroid Carcinoma (PTC) is the most common thyroid cancer, with an etiology and progression that are not fully understood. Research suggests a link between cathepsins and PTC, but the causal nature of this link is unclear. This study uses Mendelian Randomization (MR) to investigate if cathepsins causally influence PTC risk.
View Article and Find Full Text PDFRemission of Insulin-Dependent Diabetes Mellitus in Multiple Endocrine Neoplasia Type 2A After Adrenalectomy.
JCEM Case Rep
January 2025
Division of Diabetology and Metabolism, Department of Internal Medicine, Tokyo Women's Medical University School of Medicine, Shinjuku-ku, Tokyo 162-8666, Japan.
A 37-year-old man presented with symptoms of polyuria and weight loss over the past year. Initial laboratory examination showed elevated blood glucose level (468 mg/dL [25.9 mmol/L]; normal reference range [RR], 75-109 mg/dL [4.
View Article and Find Full Text PDFPerception of health and illness and quality of life after total thyroidectomy for differentiated thyroid carcinoma: the PERSAM study.
Front Endocrinol (Lausanne)
January 2025
Unit of Biostatistics, Epidemiology and Public Health, Department of Cardiac-Thoracic-Vascular Sciences and Public Health, University of Padova, Padua, Italy.
Background: Differentiated thyroid carcinoma is the most common endocrine neoplasm; several studies have shown that individuals perceive the disease as being more severe than it actually is, resulting in a reduced quality of life. The primary aim of this study is to assess the quality of life and perception of illness among patients admitted for radiometabolic therapy, post total thyroidectomy for differentiated thyroid carcinoma. The secondary aim is to identify which patient characteristics are associated with a lower quality of life in order to improve and personalize care.
View Article and Find Full Text PDFOatmeal and wheat flour as the sources of thyroid peroxidase and proinflammatory enzymes modulators in the prevention of thyroid diseases.
Sci Rep
January 2025
Department of Biochemistry and Food Chemistry, University of Life Sciences, Skromna Str. 8, Lublin, 20-704, Poland.
Polyphenolic plant compounds possess nutritional and pro-healthy potential, reducing the risk of auto-inflammatory and neoplastic diseases. However, their interference with the progression of thyroid gland dysfunctions has remained largely unaddressed. For this purpose, we combined the analyses of phenolic content and antioxidative activity with the thyroid peroxidase (TPO), lipoxygenase (LOX), xanthine oxidase (XO) and cyclooxygenase-2 (COX-2) activity assays, isobolographic approach and the estimation of thyroid cancer cells' proliferation and motility in vitro.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!