Publications by authors named "Qianmeng Pan"
Article Synopsis
- The study aimed to assess the effectiveness of ultrasound-based deep learning (DL) models in differentiating between breast phyllodes tumors (PTs) and fibroadenomas (FAs) while supporting radiologists of varying experience levels.
- Researchers collected 1180 ultrasound images from 539 patients and trained five different DL models, finding that the Xception model achieved the highest diagnostic performance, surpassing all radiologists in accuracy.
- The DL model not only showed better predictive capabilities than experienced radiologists but also improved the accuracy of diagnosis for radiologists with different levels of expertise by up to 4%.
Article Synopsis
- A deep learning model was developed to analyze ultrasound images for accurately distinguishing between benign and malignant parotid tumors, aiming to assist clinicians in diagnosis.
- The study involved 2,211 ultrasound images from 980 confirmed cases, and the Resnet18 model outperformed others with high AUC scores (0.947 for internal tests and 0.925 for external tests) as well as strong accuracy, sensitivity, and specificity.
- Radiologists' diagnostic performance improved when assisted by the model, with both junior and senior radiologists showing increased AUC values, indicating the model's potential to enhance clinical decision-making.
Background: Calcification is a common phenomenon in both benign and malignant thyroid nodules. However, the clinical significance of calcification remains unclear. Therefore, we explored a more objective method for distinguishing between benign and malignant thyroid calcified nodules.
View Article and Find Full Text PDFValue of Artificial Intelligence in Improving the Accuracy of Diagnosing TI-RADS Category 4 Nodules.
Ultrasound Med Biol
November 2023
Objective: Considerable heterogeneity is observed in the malignancy rates of thyroid nodules classified as category 4 according to the Thyroid Imaging Reporting and Data System (TI-RADS). This study was aimed at comparing the diagnostic performance of artificial intelligence algorithms and radiologists with different experience levels in distinguishing benign and malignant TI-RADS 4 (TR4) nodules.
Methods: Between January 2019 and September 2022, 1117 TR4 nodules with well-defined pathological findings were collected for this retrospective study.
Background: Both calcification and colloid in thyroid nodules are reflected as echogenic foci in ultrasound images. However, calcification and colloid have significantly different probabilities of malignancy. We explored the performance of a deep learning (DL) model in distinguishing the echogenic foci of thyroid nodules as calcification or colloid.
View Article and Find Full Text PDFObjective: The aim of this study is to develop AI-assisted software incorporating a deep learning (DL) model based on static ultrasound images. The software aims to aid physicians in distinguishing between malignant and benign thyroid nodules with echogenic foci and to investigate how the AI-assisted DL model can enhance radiologists' diagnostic performance.
Methods: For this retrospective study, a total of 2724 ultrasound (US) scans were collected from two independent institutions, encompassing 1038 echogenic foci nodules.