Enhancing the reliability and accuracy of AI-enabled diagnosis via complementarity-driven deferral to clinicians

Article Multidisciplinary Sciences

Early user experience and lessons learned using ultra-portable digital X-ray with computer-aided detection (DXR-CAD) products: A qualitative study from the perspective of healthcare providers

Zhi Zhen Qin et al.

Summary: This study assesses the experiences and lessons learned from early implementers who used ultra-portable digital X-ray systems integrated with computer-aided detection software to screen and triage TB patients. The study found that the portability of the system was a major advantage, but there were also compromises in terms of throughput, internet dependence, manoeuvrability, stability, and suitability for larger patients.

PLOS ONE (2023)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

Artificial Intelligence Evaluation of 122 969 Mammography Examinations from a Population-based Screening Program

Marthe Larsen et al.

Summary: This study compares the performance of a commercially available AI system with routine double reading in breast cancer detection. The AI system shows a high detection rate and the histopathologic characteristics of tumors with different AI scores are explored, providing valuable insights into the screening outcomes.

RADIOLOGY (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

An interpretable classifier for high-resolution breast cancer screening images utilizing weakly supervised localization

Yiqiu Shen et al.

Summary: A novel neural network model is proposed for medical image analysis, addressing the unique properties of medical images. The model is trained with image-level labels only and can generate pixel-level saliency maps indicating possible malignant findings. The model outperforms existing methods and achieves high accuracy while being computationally efficient.

MEDICAL IMAGE ANALYSIS (2021)

Add to Collection

Review Medicine, General & Internal

Use of artificial intelligence for image analysis in breast cancer screening programmes: systematic review of test accuracy

Karoline Freeman et al.

Summary: This study aimed to examine the accuracy of artificial intelligence in breast cancer mammography screening practice. The evidence does not yet support the use of AI to replace the double reading of radiologists in screening programs, and further prospective studies are needed to determine the potential benefits of AI in clinical practice.

BMJ-BRITISH MEDICAL JOURNAL (2021)

Add to Collection

Article Health Care Sciences & Services

Do as AI say: susceptibility in deployment of clinical decision-aids

Susanne Gaube et al.

Summary: This study found that radiologists rated diagnostic advice as lower quality when it appeared to come from an AI system, while less experienced physicians did not have this bias. Diagnostic accuracy significantly decreased when participants received inaccurate advice, regardless of the purported source being AI or human experts. Therefore, important considerations need to be made when deploying advice in clinical environments, whether it is coming from AI or non-AI sources.

NPJ DIGITAL MEDICINE (2021)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Big Self-Supervised Models Advance Medical Image Classification

Shekoofeh Azizi et al.

Summary: Self-supervised pretraining followed by supervised fine-tuning has been successful in medical image classification, significantly improving accuracy. The Multi-Instance Contrastive Learning method shows advantages in constructing informative positive pairs for self-supervised learning. Big self-supervised models are robust to distribution shift and can efficiently learn even with a small number of labeled medical images.

2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021) (2021)

Add to Collection

Article Multidisciplinary Sciences