Video from SIIM 2023: Marc Kohli on validating AI

Jun 15, 2023
2023 06 20 15 19 1037 Kohli Marc 400

AUSTIN, TX - How can radiology practices trust the AI technology they want to implement? Marc Kohli, MD, of the University of California, San Francisco shared his perspectives on the validation and monitoring of AI at the annual meeting of the Society for Imaging Informatics in Medicine (SIIM).


Latest in AI
Images Large Radiol Thumbnail
Generative AI model shows promise in chest x-ray study
March 25, 2025
Emergency
ML-based system predicts x-ray service times
March 20, 2025
Illusion Thumbnail
Radiologists less susceptible to visual illusions
March 20, 2025
Full-field digital screening mammograms in participants recalled by the consensus discussion after being flagged by artificial intelligence (AI) computer-aided detection (CAD) but not by either of the two radiologists. (A) Mammogram with AI CAD score of 73 for a 55-year-old woman diagnosed with grade 3 in situ cancer and T1 (9-mm) invasive cancer in the left breast and lymph node metastasis. (B) Mammogram with AI CAD score of 54 for a 49-year-old woman diagnosed with grade 2 in situ cancer and T1 (6-mm) invasive cancer in the right breast and without lymph node metastasis. All images and caption courtesy of the RSNA.
Do radiologists trust mammography AI software?
March 18, 2025
Related Stories
2019 12 09 17 23 6730 Artificial Intelligence Ai Suit 400
Advanced Visualization
Marketing, regulatory summaries may differ with AI tech
2023 06 15 19 25 6614 Kim Woojin 400
AI
Video from SIIM 2023: There's more to generative AI than ChatGPT
2017 04 20 14 33 37 718 Money Hundred Bills 400
AI
Medivis raises $20M in funding round
2021 04 07 22 09 9651 Artificial Intelligence Ai Face 400
AI
SIIM: Imaging AI developers need meaningful training data
More in AI
Generative AI model shows promise in chest x-ray study
By Amerigo Allegretto
A multimodal generative AI model achieved high diagnostic accuracy and showed clinical value in preliminary reporting of chest x-ray images.
March 25, 2025
Images Large Radiol Thumbnail
ML-based system predicts x-ray service times
By Will Morton
A team in Italy has developed a real-time machine-learning (ML) system that can predict x-ray service times in emergency departments.
March 20, 2025
Emergency
Radiologists less susceptible to visual illusions
By Will Morton
Medical image experts demonstrated superior perceptual accuracy in response to visual illusions compared with a control group.
March 20, 2025
Illusion Thumbnail
Do radiologists trust mammography AI software?
By Erik L. Ridley
A large-scale prospective trial exploring the use of mammography AI software in a breast cancer screening program has found that recalls were lower when cases were flagged by AI.
March 18, 2025
Full-field digital screening mammograms in participants recalled by the consensus discussion after being flagged by artificial intelligence (AI) computer-aided detection (CAD) but not by either of the two radiologists. (A) Mammogram with AI CAD score of 73 for a 55-year-old woman diagnosed with grade 3 in situ cancer and T1 (9-mm) invasive cancer in the left breast and lymph node metastasis. (B) Mammogram with AI CAD score of 54 for a 49-year-old woman diagnosed with grade 2 in situ cancer and T1 (6-mm) invasive cancer in the right breast and without lymph node metastasis. All images and caption courtesy of the RSNA.
DL algorithm targeting prostate cancer demonstrates clinical impact
By Liz Carey
Per participant, a deep-learning (DL) algorithm demonstrated increased specificity of 64.8% compared with radiologists at 21.3%.
March 12, 2025
MRI scans show two areas of clinically significant prostate cancer in a 65-year-old man with a prostate-specific antigen level of 8.23 ng/mL. Images are (A, E) axial T2-weighted images, (B, F) axial diffusion-weighted images obtained with a b value of 1500 sec/mm2, (C, G) axial apparent diffusion coefficient images, and (D, H) lesion masks generated by the deep learning algorithm (DLA) on T2-weighted images. The DLA detected a lesion in the right posteromedial peripheral zone of the apex (D), which was assigned as Prostate Imaging Reporting and Data System (PI-RADS) 4. A radiologist also detected the lesion and assigned it PI-RADS 4 (arrow in A-C). A lesion in the right anterior transition zone (H) was detected by the DLA, which was assigned PI-RADS 4. The radiologist did not detect the lesion (arrow in E-G). Both lesions were confirmed as prostate cancer with Gleason grade group 2. Caption and image courtesy of RSNA.
AI boosts identification of incident vertebral fractures on chest CT
By Kate Madden Yee
An AI algorithm that measures 3D whole thoracic vertebral bone mineral density on conventional chest CT can help predict incident vertebral fractures.
March 11, 2025
Back Spine Skeleton
Deep-learning radiomics model aids in ovarian cancer diagnosis
By Amerigo Allegretto
An ensemble model combining clinical variables, O-RADS, and deep learning radiomics can diagnose ovarian tumors.
March 5, 2025
(A) SHAP summary plot shows the global contributions of variables to the model predictions. (B-E) Local explanation of the model for a single instance of ovarian endometrioid carcinoma: SHAP waterfall plot (B), ultrasound grayscale image (C), ultrasound color Doppler image (D), and 10×10 histopathological image (E). Images are available for republishing under a Creative Commons license (CC BY-NC-ND 4.0).
AI model aimed at adrenal glands, nodules could reduce workload by 10%
By Liz Carey
A future vision of AI in this context could entail analyzing adrenal nodules across multiple phases, extracting radiomic data, and providing a predictability score, study authors said.
March 4, 2025
Images Large Radiol Thumbnail
CNN models help determine best treatment for kids with GI obstructions
By Kate Madden Yee
Convolutional neural network (CNN) algorithms show promise for identifying gastrointestinal (GI) obstructions on x-ray in a pediatric population, researchers have reported.
February 28, 2025
Digestive System 3 D Intestine
ECR: Integrating AI into radiology workflows benefits the environment
By Kate Madden Yee
Sustainability in radiology is a hot topic at this year's ECR in Vienna, and a February 27 talk focused on "green" radiology using AI technology.
February 28, 2025
Am 2020 03 04 21 21 6197 Artificial Intelligence Ai Hand 400 20200304215813
ECR: Repurposed AI boosts breast cancer detection in diverse patient groups
By Amerigo Allegretto
Repurposed AI imaging tools can improve breast malignancy detection across diverse patient groups.
February 27, 2025
Artificial Intelligence Ai
ECR: Generative AI yields potential for assisting radiologists
By Amerigo Allegretto
Generative AI may have novel applications in the clinic, but using the technology comes with its share of challenges and future directions.
February 26, 2025
Marc Kohli, MD, discusses current and potential future uses of generative AI in radiology at ECR 2025 in Vienna, Austria. Behind him is an image created by one such generative AI model, which illustrates humans working alongside AI.
Page 1 of 375
Next Page