Retrospective Review of Missed Cancer Detection and Its Mammography Findings with Artificial-Intelligence-Based, Computer-Aided Diagnosis

Artificial Intelligence for Breast Cancer Detection on Mammography: Factors Related to Cancer Detection

RATIONALE AND OBJECTIVES

Little is known about the factors affecting the Artificial Intelligence (AI) software performance on mammography for breast cancer detection. This study was to identify factors associated with abnormality scores assigned by the AI software.

MATERIALS AND METHODS

A retrospective database search was conducted to identify consecutive asymptomatic women who underwent breast cancer surgery between April 2016 and December 2019. A commercially available AI software (Lunit INSIGHT, MMG, Ver. 1.1.4.0) was used for preoperative mammography to assign individual abnormality scores to the lesions and score of 10 or higher was considered as positive detection by AI software. Radiologists without knowledge of the AI results retrospectively assessed the mammographic density and classified mammographic findings into positive and negative finding. General linear model (GLM) analysis was used to identify the clinical, pathological, and mammographic findings related to the abnormality scores, obtaining coefficient β values that represent the mean difference per unit or comparison with the reference value. Additionally, the reasons for non-detection by the AI software were investigated.

RESULTS

Among the 1001 index cancers (830 invasive cancers and 171 ductal carcinoma in situs) in 1001 patients, 717 (72%) were correctly detected by AI, while the remaining 284 (28%) were not detected. Multivariable GLM analysis showed that abnormal mammography findings (β = 77.0 for mass, β = 73.1 for calcification only, β = 49.4 for architectural distortion, and β = 47.6 for asymmetry compared to negative; all Ps < 0.001), invasive tumor size (β = 4.3 per 1 cm, P < 0.001), and human epidermal growth receptor type 2 (HER2) positivity (β = 9.2 compared to hormone receptor positive, HER2 negative, P = 0.004) were associated with higher mean abnormality score. AI failed to detect small asymmetries in extremely dense breasts, subcentimeter-sized or isodense lesions, and faint amorphous calcifications.

CONCLUSION

Cancers with positive abnormal mammographic findings on retrospective review, large invasive size, HER2 positivity had high AI abnormality scores. Understanding the patterns of AI software performance is crucial for effectively integrating AI into clinical practice.

Aptamer-functionalized MOFs and AI-driven strategies for early cancer diagnosis and therapeutics

Metal-Organic Frameworks (MOFs) have exceptional inherent properties that make them highly suitable for diverse applications, such as catalysis, storage, optics, chemo sensing, and biomedical science and technology. Over the past decades, researchers have utilized various techniques, including solvothermal, hydrothermal, mechanochemical, electrochemical, and ultrasonic, to synthesize MOFs with tailored properties. Post-synthetic modification of linkers, nodal components, and crystallite domain size and morphology can functionalize MOFs to improve their aptamer applications. Advancements in AI and machine learning led to the development of nonporous MOFs and nanoscale MOFs for medical purposes. MOFs have exhibited promise in cancer therapy, with the successful accumulation of a photosensitizer in cancer cells representing a significant breakthrough. This perspective is focused on MOFs' use as advanced materials and systems for cancer therapy, exploring the challenging aspects and promising features of MOF-based cancer diagnosis and treatment. The paper concludes by emphasizing the potential of MOFs as a transformative technology for cancer treatment and diagnosis.

AI Risk Score on Screening Mammograms Preceding Breast Cancer Diagnosis

Background Few studies have evaluated the role of artificial intelligence (AI) in prior screening mammography. Purpose To examine AI risk scores assigned to screening mammography in women who were later diagnosed with breast cancer. Materials and Methods Image data and screening information of examinations performed from January 2004 to December 2019 as part of BreastScreen Norway were used in this retrospective study. Prior screening examinations from women who were later diagnosed with cancer were assigned an AI risk score by a commercially available AI system (scores of 1-7, low risk of malignancy; 8-9, intermediate risk; and 10, high risk of malignancy). Mammographic features of the cancers based on the AI score were also assessed. The association between AI score and mammographic features was tested with a bivariate test. Results A total of 2787 prior screening examinations from 1602 women (mean age, 59 years ± 5.1 [SD]) with screen-detected (n = 1016) or interval (n = 586) cancers showed an AI risk score of 10 for 389 (38.3%) and 231 (39.4%) cancers, respectively, on the mammograms in the screening round prior to diagnosis. Among the screen-detected cancers with AI scores available two screening rounds (4 years) before diagnosis, 23.0% (122 of 531) had a score of 10. Mammographic features were associated with AI score for invasive screen-detected cancers (P < .001). Density with calcifications was registered for 13.6% (43 of 317) of screen-detected cases with a score of 10 and 4.6% (15 of 322) for those with a score of 1-7. Conclusion More than one in three cases of screen-detected and interval cancers had the highest AI risk score at prior screening, suggesting that the use of AI in mammography screening may lead to earlier detection of breast cancers. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Mehta in this issue.

Artificial Intelligence Improves Detection of Supplemental Screening Ultrasound-detected Breast Cancers in Mammography

Despite recent advances in artificial intelligence (AI) software with improved performance in mammography screening for breast cancer, insufficient data are available on its performance in detecting cancers that were initially missed on mammography. In this study, we aimed to determine whether AI software-aided mammography could provide additional value in identifying cancers detected through supplemental screening ultrasound. We searched our database from 2017 to 2018 and included 238 asymptomatic patients (median age, 50 years; interquartile range, 45-57 years) diagnosed with breast cancer using supplemental ultrasound. Two unblinded radiologists retrospectively reviewed the mammograms using commercially available AI software and identified the reasons for missed detection. Clinicopathological characteristics of AI-detected and AI-undetected cancers were compared using univariate and multivariate logistic regression analyses. A total of 253 cancers were detected in 238 patients using ultrasound. In an unblinded review, the AI software failed to detect 187 of the 253 (73.9%) mammography cases with negative findings in retrospective observations. The AI software detected 66 cancers (26.1%), of which 42 (63.6%) exhibited indiscernible findings obscured by overlapping dense breast tissues, even with the knowledge of magnetic resonance imaging and post-wire localization mammography. The remaining 24 cases (36.4%) were considered interpretive errors by the radiologists. Invasive tumor size was associated with AI detection after multivariable analysis (odds ratio, 2.2; 95% confidence intervals, 1.5-3.3; p < 0.001). In the control group of 160 women without cancer, the AI software identified 19 false positives (11.9%, 19/160). Although most ultrasound-detected cancers were not detected on mammography with the use of AI, the software proved valuable in identifying breast cancers with indiscernible abnormalities or those that clinicians may have overlooked.

BUViTNet: Breast Ultrasound Detection via Vision Transformers

Convolutional neural networks (CNNs) have enhanced ultrasound image-based early breast cancer detection. Vision transformers (ViTs) have recently surpassed CNNs as the most effective method for natural image analysis. ViTs have proven their capability of incorporating more global information than CNNs at lower layers, and their skip connections are more powerful than those of CNNs, which endows ViTs with superior performance. However, the effectiveness of ViTs in breast ultrasound imaging has not yet been investigated. Here, we present BUViTNet breast ultrasound detection via ViTs, where ViT-based multistage transfer learning is performed using ImageNet and cancer cell image datasets prior to transfer learning for classifying breast ultrasound images. We utilized two publicly available ultrasound breast image datasets, Mendeley and breast ultrasound images (BUSI), to train and evaluate our algorithm. The proposed method achieved the highest area under the receiver operating characteristics curve (AUC) of 1 ± 0, Matthew’s correlation coefficient (MCC) of 1 ± 0, and kappa score of 1 ± 0 on the Mendeley dataset. Furthermore, BUViTNet achieved the highest AUC of 0.968 ± 0.02, MCC of 0.961 ± 0.01, and kappa score of 0.959 ± 0.02 on the BUSI dataset. BUViTNet outperformed ViT trained from scratch, ViT-based conventional transfer learning, and CNN-based transfer learning in classifying breast ultrasound images (p < 0.01 in all cases). Our findings indicate that improved transformers are effective in analyzing breast images and can provide an improved diagnosis if used in clinical settings. Future work will consider the use of a wide range of datasets and parameters for optimized performance.

Clinical deployment environments: Five pillars of translational machine learning for health

Machine Learning for Health (ML4H) has demonstrated efficacy in computer imaging and other self-contained digital workflows, but has failed to substantially impact routine clinical care. This is no longer because of poor adoption of Electronic Health Records Systems (EHRS), but because ML4H needs an infrastructure for development, deployment and evaluation within the healthcare institution. In this paper, we propose a design pattern called a Clinical Deployment Environment (CDE). We sketch the five pillars of the CDE: (1) real world development supported by live data where ML4H teams can iteratively build and test at the bedside (2) an ML-Ops platform that brings the rigour and standards of continuous deployment to ML4H (3) design and supervision by those with expertise in AI safety (4) the methods of implementation science that enable the algorithmic insights to influence the behaviour of clinicians and patients and (5) continuous evaluation that uses randomisation to avoid bias but in an agile manner. The CDE is intended to answer the same requirements that bio-medicine articulated in establishing the translational medicine domain. It envisions a transition from "real-world" data to "real-world" development.