Artificial Intelligence in Breast Cancer Screening: Evaluation of FDA Device Regulation and Future Recommendations

XRAInet: AI-based decision support for pneumothorax and pleural effusion management

PURPOSE

This study aimed to develop and assess the performance of an artificial intelligence (AI)-driven decision support system, XRAInet, in accurately identifying pediatric patients with pleural effusion or pneumothorax and determining whether tube thoracostomy intervention is warranted.

METHODS

In this diagnostic accuracy study, we retrospectively analyzed a data set containing 510 X-ray images from 170 pediatric patients admitted between 2005 and 2022. Patients were categorized into two groups: Tube (requiring tube thoracostomy) and Conservative (managed conservatively). XRAInet, a deep learning-based algorithm, was trained using this data set. We evaluated its performance using various metrics, including mean Average Precision (mAP), recall, precision, and F1 score.

RESULTS

XRAInet, achieved a mAP score of 0.918. This result underscores its ability to accurately identify and localize regions necessitating tube thoracostomy for pediatric patients with pneumothorax and pleural effusion. In an independent testing data set, the model exhibited a sensitivity of 64.00% and specificity of 96.15%.

CONCLUSION

In conclusion, XRAInet presents a promising solution for improving the detection and decision-making process for cases of pneumothorax and pleural effusion in pediatric patients using X-ray images. These findings contribute to the expanding field of AI-driven medical imaging, with potential applications for enhancing patient outcomes. Future research endeavors should explore hybrid models, enhance interpretability, address data quality issues, and align with regulatory requirements to ensure the safe and effective deployment of XRAInet in healthcare settings.

Clinical Integration of Artificial Intelligence for Breast Imaging

This article describes an approach to planning and implementing artificial intelligence products in a breast screening service. It highlights the importance of an in-depth understanding of the end-to-end workflow and effective project planning by a multidisciplinary team. It discusses the need for monitoring to ensure that performance is stable and meets expectations, as well as focusing on the potential for inadvertantly generating inequality. New cross-discipline roles and expertise will be needed to enhance service delivery.

Screening for liver fibrosis: lessons from colorectal and lung cancer screening

Many countries have incorporated population screening programmes for cancer, such as colorectal and lung cancer, into their health-care systems. Cirrhosis is more prevalent than colorectal cancer and has a comparable age-standardized mortality rate to lung cancer. Despite this fact, there are no screening programmes in place for early detection of liver fibrosis, the precursor of cirrhosis. In this Perspective, we use insights from colorectal and lung cancer screening to explore the benefits, challenges, implementation strategies and pathways for future liver fibrosis screening initiatives. Several non-invasive methods and referral pathways for early identification of liver fibrosis exist, but in addition to accurate detection, screening programmes must also be cost-effective and demonstrate benefit through a reduction in liver-related mortality. Randomized controlled trials are needed to confirm this. Future randomized screening trials should evaluate not only the screening tests, but also interventions used to halt disease progression in individuals identified through screening.

Familiarity, confidence and preference of artificial intelligence feedback and prompts by Australian breast cancer screening readers

Objectives This study explored the familiarity, perceptions and confidence of Australian radiology clinicians involved in reading screening mammograms, regarding artificial intelligence (AI) applications in breast cancer detection. Methods Sixty-five radiologists, breast physicians and radiology trainees participated in an online survey that consisted of 23 multiple choice questions asking about their experience and familiarity with AI products. Furthermore, the survey asked about their confidence in using AI outputs and their preference for AI modes applied in a breast screening context. Participants' responses to questions were compared using Pearson's χ 2 test. Bonferroni-adjusted significance tests were used for pairwise comparisons. Results Fifty-five percent of respondents had experience with AI in their workplaces, with automatic density measurement powered by machine learning being the most familiar AI product (69.4%). The top AI outputs with the highest ranks of perceived confidence were 'Displaying suspicious areas on mammograms with the percentage of cancer possibility' (67.8%) and 'Automatic mammogram classification (normal, benign, cancer, uncertain)' (64.6%). Radiology and breast physicians preferred using AI as second-reader mode (75.4% saying 'somewhat happy' to 'extremely happy') over triage (47.7%), pre-screening and first-reader modes (both with 26.2%) (P  < 0.001). Conclusion The majority of screen readers expressed increased confidence in utilising AI for highlighting suspicious areas on mammograms and for automatically classifying mammograms. They considered AI as an optimal second-reader mode being the most ideal use in a screening program. The findings provide valuable insights into the familiarities and expectations of radiologists and breast clinicians for the AI products that can enhance the effectiveness of the breast cancer screening programs, benefitting both healthcare professionals and patients alike.

Artificial intelligence assisted diagnosis of early tc markers and its application

Thyroid cancer (TC) is a common endocrine malignancy with an increasing incidence worldwide. Early diagnosis is particularly important for TC patients, because it allows patients to receive treatment as early as possible. Artificial intelligence (AI) provides great advantages for complex healthcare systems by analyzing big data based on machine learning. Nowadays, AI is widely used in the early diagnosis of cancer such as TC. Ultrasound detection and fine needle aspiration biopsy are the main methods for early diagnosis of TC. AI has been widely used in the detection of malignancy in thyroid nodules by ultrasound images, cytopathology images and molecular markers. It shows great potential in auxiliary medical diagnosis. The latest clinical trial has shown that the performance of AI models matches with the diagnostic efficiency of experienced clinicians, and more efficient AI tools will be developed in the future. Therefore, in this review, we summarized the recent advances in the application of AI algorithms in assessing the risk of malignancy in thyroid nodules. The objective of this review was to provide a data base for the clinical use of AI-assisted diagnosis in TC, as well as to provide new ideas for the next generation of AI-assisted diagnosis in TC.

Evaluating Magnetic Seed Localization in Targeted Axillary Dissection for Node-Positive Early Breast Cancer Patients Receiving Neoadjuvant Systemic Therapy: A Comprehensive Review and Pooled Analysis

Background/Objectives: De-escalation of axillary surgery is made possible by advancements in both neoadjuvant systemic therapy (NST) and in localisation technology for breast lesions. Magseed®, developed in 2013 by Dr. Michael Douk of Cambridge, United Kingdom, is a wire-free localisation technology that facilitates the localisation and retrieval of lymph nodes for staging. Targeted axillary dissection (TAD), which entails marked lymph node biopsy (MLNB) and sentinel lymph node biopsy (SLNB), has emerged as the preferred method to assess residual disease in post-NST node-positive patients. This systematic review and pooled analysis evaluate the performance of Magseed® in TAD. Methods: The search was carried out in PubMed and Google Scholar. An assessment of localisation, retrieval rates, concordance between MLNB and SLNB, and pathological complete response (pCR) in clinically node-positive patients post NST was undertaken. Results: Nine studies spanning 494 patients and 497 procedures were identified, with a 100% successful deployment rate, a 94.2% (468/497) [95% confidence interval (CI), 93.7-94.7] localisation rate, a 98.8% (491/497) retrieval rate, and a 68.8% (247/359) [95% CI 65.6-72.0] concordance rate. pCR was observed in 47.9% (220/459) ) [95% CI 43.3-52.6] of cases. Subgroup analysis of studies reporting the pathological status of MLNB and SLNB separately revealed an FNR of 4.2% for MLNB and 17.6% for SLNB (p = 0.0013). Mean duration of implantation was 37 days (range: 0-188). Conclusions: These findings highlight magnetic seed localisation's efficacy in TAD for NST-treated node-positive patients, aiding in accurate axillary pCR identification and safe de-escalation of axillary surgery in excellent responders.

Patient Characteristics Impact Performance of AI Algorithm in Interpreting Negative Screening Digital Breast Tomosynthesis Studies

Background Artificial intelligence (AI) is increasingly used to manage radiologists' workloads. The impact of patient characteristics on AI performance has not been well studied. Purpose To understand the impact of patient characteristics (race and ethnicity, age, and breast density) on the performance of an AI algorithm interpreting negative screening digital breast tomosynthesis (DBT) examinations. Materials and Methods This retrospective cohort study identified negative screening DBT examinations from an academic institution from January 1, 2016, to December 31, 2019. All examinations had 2 years of follow-up without a diagnosis of atypia or breast malignancy and were therefore considered true negatives. A subset of unique patients was randomly selected to provide a broad distribution of race and ethnicity. DBT studies in this final cohort were interpreted by a U.S. Food and Drug Administration-approved AI algorithm, which generated case scores (malignancy certainty) and risk scores (1-year subsequent malignancy risk) for each mammogram. Positive examinations were classified based on vendor-provided thresholds for both scores. Multivariable logistic regression was used to understand relationships between the scores and patient characteristics. Results A total of 4855 patients (median age, 54 years [IQR, 46-63 years]) were included: 27% (1316 of 4855) White, 26% (1261 of 4855) Black, 28% (1351 of 4855) Asian, and 19% (927 of 4855) Hispanic patients. False-positive case scores were significantly more likely in Black patients (odds ratio [OR] = 1.5 [95% CI: 1.2, 1.8]) and less likely in Asian patients (OR = 0.7 [95% CI: 0.5, 0.9]) compared with White patients, and more likely in older patients (71-80 years; OR = 1.9 [95% CI: 1.5, 2.5]) and less likely in younger patients (41-50 years; OR = 0.6 [95% CI: 0.5, 0.7]) compared with patients aged 51-60 years. False-positive risk scores were more likely in Black patients (OR = 1.5 [95% CI: 1.0, 2.0]), patients aged 61-70 years (OR = 3.5 [95% CI: 2.4, 5.1]), and patients with extremely dense breasts (OR = 2.8 [95% CI: 1.3, 5.8]) compared with White patients, patients aged 51-60 years, and patients with fatty density breasts, respectively. Conclusion Patient characteristics influenced the case and risk scores of a Food and Drug Administration-approved AI algorithm analyzing negative screening DBT examinations. © RSNA, 2024.

Research ethics and artificial intelligence for global health: perspectives from the global forum on bioethics in research

BACKGROUND

The ethical governance of Artificial Intelligence (AI) in health care and public health continues to be an urgent issue for attention in policy, research, and practice. In this paper we report on central themes related to challenges and strategies for promoting ethics in research involving AI in global health, arising from the Global Forum on Bioethics in Research (GFBR), held in Cape Town, South Africa in November 2022.

METHODS

The GFBR is an annual meeting organized by the World Health Organization and supported by the Wellcome Trust, the US National Institutes of Health, the UK Medical Research Council (MRC) and the South African MRC. The forum aims to bring together ethicists, researchers, policymakers, research ethics committee members and other actors to engage with challenges and opportunities specifically related to research ethics. In 2022 the focus of the GFBR was "Ethics of AI in Global Health Research". The forum consisted of 6 case study presentations, 16 governance presentations, and a series of small group and large group discussions. A total of 87 participants attended the forum from 31 countries around the world, representing disciplines of bioethics, AI, health policy, health professional practice, research funding, and bioinformatics. In this paper, we highlight central insights arising from GFBR 2022.

RESULTS

We describe the significance of four thematic insights arising from the forum: (1) Appropriateness of building AI, (2) Transferability of AI systems, (3) Accountability for AI decision-making and outcomes, and (4) Individual consent. We then describe eight recommendations for governance leaders to enhance the ethical governance of AI in global health research, addressing issues such as AI impact assessments, environmental values, and fair partnerships.

CONCLUSIONS

The 2022 Global Forum on Bioethics in Research illustrated several innovations in ethical governance of AI for global health research, as well as several areas in need of urgent attention internationally. This summary is intended to inform international and domestic efforts to strengthen research ethics and support the evolution of governance leadership to meet the demands of AI in global health research.

Can multi-modal radiomics using pretreatment ultrasound and tomosynthesis predict response to neoadjuvant systemic treatment in breast cancer?

OBJECTIVES

Response assessment to neoadjuvant systemic treatment (NAST) to guide individualized treatment in breast cancer is a clinical research priority. We aimed to develop an intelligent algorithm using multi-modal pretreatment ultrasound and tomosynthesis radiomics features in addition to clinical variables to predict pathologic complete response (pCR) prior to the initiation of therapy.

METHODS

We used retrospective data on patients who underwent ultrasound and tomosynthesis before starting NAST. We developed a support vector machine algorithm using pretreatment ultrasound and tomosynthesis radiomics features in addition to patient and tumor variables to predict pCR status (ypT0 and ypN0). Findings were compared to the histopathologic evaluation of the surgical specimen. The main outcome measures were area under the curve (AUC) and false-negative rate (FNR).

RESULTS

We included 720 patients, 504 in the development set and 216 in the validation set. Median age was 51.6 years and 33.6% (242 of 720) achieved pCR. The addition of radiomics features significantly improved the performance of the algorithm (AUC 0.72 to 0.81; p = 0.007). The FNR of the multi-modal radiomics and clinical algorithm was 6.7% (10 of 150 with missed residual cancer). Surface/volume ratio at tomosynthesis and peritumoral entropy characteristics at ultrasound were the most relevant radiomics. Hormonal receptors and HER-2 status were the most important clinical predictors.

CONCLUSION

A multi-modal machine learning algorithm with pretreatment clinical, ultrasound, and tomosynthesis radiomics features may aid in predicting residual cancer after NAST. Pending prospective validation, this may facilitate individually tailored NAST regimens.

CLINICAL RELEVANCE STATEMENT

Multi-modal radiomics using pretreatment ultrasound and tomosynthesis showed significant improvement in assessing response to NAST compared to an algorithm using clinical variables only. Further prospective validation of our findings seems warranted to enable individualized predictions of NAST outcomes.

KEY POINTS

• We proposed a multi-modal machine learning algorithm with pretreatment clinical, ultrasound, and tomosynthesis radiomics features to predict response to neoadjuvant breast cancer treatment. • Compared with the clinical algorithm, the AUC of this integrative algorithm is significantly higher. • Used prior to the initiative of therapy, our algorithm can identify patients who will experience pathologic complete response following neoadjuvant therapy with a high negative predictive value.

Ultrasound Radiomics Features to Identify Patients With Triple-Negative Breast Cancer: A Retrospective, Single-Center Study

OBJECTIVES

Patients with triple-negative breast cancer (TNBC) exhibit a fast tumor growth rate and poor survival outcomes. In this study, we aimed to develop and compare intelligent algorithms using ultrasound radiomics features in addition to clinical variables to identify patients with TNBC prior to histopathologic diagnosis.

METHODS

We used single-center, retrospective data of patients who underwent ultrasound before histopathologic verification and subsequent neoadjuvant systemic treatment (NAST). We developed a logistic regression with an elastic net penalty algorithm using pretreatment ultrasound radiomics features in addition to patient and tumor variables to identify patients with TNBC. Findings were compared to the histopathologic evaluation of the biopsy specimen. The main outcome measure was the area under the curve (AUC).

RESULTS

We included 1161 patients, 813 in the development set and 348 in the validation set. Median age was 50.1 years and 24.4% (283 of 1161) had TNBC. The integrative model using radiomics and clinical information showed significantly better performance in identifying TNBC compared to the radiomics model (AUC: 0.71, 95% confidence interval [CI]: 0.65-0.76 versus 0.64, 95% CI: 0.57-0.71, P = .004). The five most important variables were cN status, shape surface volume ratio (SA:V), gray level co-occurrence matrix (GLCM) correlation, gray level dependence matrix (GLDM) dependence nonuniformity normalized, and age. Patients with TNBC were more often categorized as BI-RADS 4 than BI-RADS 5 compared to non-TNBC patients (P = .002).

CONCLUSION

A machine learning algorithm showed promising potential to identify patients with TNBC using ultrasound radiomics features and clinical information prior to histopathologic evaluation.

Detailed Image Data Quality and Cleaning Practices for Artificial Intelligence Tools for Breast Cancer

Standardizing image-data preparation practices to improve accuracy/consistency of AI diagnostic tools.

Artificial Intelligence-Powered Mammography: Navigating the Landscape of Deep Learning for Breast Cancer Detection

Worldwide, breast cancer (BC) is one of the most commonly diagnosed malignancies in women. Early detection is key to improving survival rates and health outcomes. This literature review focuses on how artificial intelligence (AI), especially deep learning (DL), can enhance the ability of mammography, a key tool in BC detection, to yield more accurate results. Artificial intelligence has shown promise in reducing diagnostic errors and increasing early cancer detection chances. Nevertheless, significant challenges exist, including the requirement for large amounts of high-quality data and concerns over data privacy. Despite these hurdles, AI and DL are advancing the field of radiology, offering better ways to diagnose, detect, and treat diseases. The U.S. Food and Drug Administration (FDA) has approved several AI diagnostic tools. Yet, the full potential of these technologies, especially for more advanced screening methods like digital breast tomosynthesis (DBT), depends on further clinical studies and the development of larger databases. In summary, this review highlights the exciting potential of AI in BC screening. It calls for more research and validation to fully employ the power of AI in clinical practice, ensuring that these technologies can help save lives by improving diagnosis accuracy and efficiency.

Breast cancer detection using enhanced IRI-numerical engine and inverse heat transfer modeling: model description and clinical validation

Effective treatment of breast cancer relies heavily on early detection. Routine annual mammography is a widely accepted screening technique that has resulted in significantly improving the survival rate. However, it suffers from low sensitivity resulting in high false positives from screening. To overcome this problem, adjunctive technologies such as ultrasound are employed on about 10% of women recalled for additional screening following mammography. These adjunctive techniques still result in a significant number of women, about 1.6%, who undergo biopsy while only 0.4% of women screened have cancers. The main reason for missing cancers during mammography screening arises from the masking effect of dense breast tissue. The presence of a tumor results in the alteration of temperature field in the breast, which is not influenced by the tissue density. In the present paper, the IRI-Numerical Engine is presented as an adjunct for detecting cancer from the surface temperature data. It uses a computerized inverse heat transfer approach based on Pennes's bioheat transfer equations. Validation of this enhanced algorithm is conducted on twenty-three biopsy-proven breast cancer patients after obtaining informed consent under IRB protocol. The algorithm correctly predicted the size and location of cancerous tumors in twenty-four breasts, while twenty-two contralateral breasts were also correctly predicted to have no cancer (one woman had bilateral breast cancer). The tumors are seen as highly perfused and metabolically active heat sources that alter the surface temperatures that are used in heat transfer modeling. Furthermore, the results from this study with twenty-four biopsy-proven cancer cases indicate that the detection of breast cancer is not affected by breast density. This study indicates the potential of the IRI-Numerical Engine as an effective adjunct to mammography. A large scale clinical study in a statistically significant sample size is needed before integrating this approach in the current protocol.

Current state and prospects of artificial intelligence in allergy

The field of medicine is witnessing an exponential growth of interest in artificial intelligence (AI), which enables new research questions and the analysis of larger and new types of data. Nevertheless, applications that go beyond proof of concepts and deliver clinical value remain rare, especially in the field of allergy. This narrative review provides a fundamental understanding of the core concepts of AI and critically discusses its limitations and open challenges, such as data availability and bias, along with potential directions to surmount them. We provide a conceptual framework to structure AI applications within this field and discuss forefront case examples. Most of these applications of AI and machine learning in allergy concern supervised learning and unsupervised clustering, with a strong emphasis on diagnosis and subtyping. A perspective is shared on guidelines for good AI practice to guide readers in applying it effectively and safely, along with prospects of field advancement and initiatives to increase clinical impact. We anticipate that AI can further deepen our knowledge of disease mechanisms and contribute to precision medicine in allergy.

Automated diagnosis and management of follicular thyroid nodules based on the devised small-dataset interpretable foreground optimization network deep learning: a multicenter diagnostic study

BACKGROUND

Currently, follicular thyroid carcinoma (FTC) has a relatively low incidence with a lack of effective preoperative diagnostic means. To reduce the need for invasive diagnostic procedures and to address information deficiencies inherent in a small dataset, we utilized interpretable foreground optimization network deep learning to develop a reliable preoperative FTC detection system.

METHODS

In this study, a deep learning model (FThyNet) was established using preoperative ultrasound images. Data on patients in the training and internal validation cohort ( n =432) were obtained from Ruijin Hospital, China. Data on patients in the external validation cohort ( n =71) were obtained from four other clinical centers. We evaluated the predictive performance of FThyNet and its ability to generalize across multiple external centers and compared the results yielded with assessments from physicians directly predicting FTC outcomes. In addition, the influence of texture information around the nodule edge on the prediction results was evaluated.

RESULTS

FThyNet had a consistently high accuracy in predicting FTC with an area under the receiver operating characteristic curve (AUC) of 89.0% [95% CI 87.0-90.9]. Particularly, the AUC for grossly invasive FTC reached 90.3%, which was significantly higher than that of the radiologists (56.1% [95% CI 51.8-60.3]). The parametric visualization study found that those nodules with blurred edges and relatively distorted surrounding textures were more likely to have FTC. Furthermore, edge texture information played an important role in FTC prediction with an AUC of 68.3% [95% CI 61.5-75.5], and highly invasive malignancies had the highest texture complexity.

CONCLUSION

FThyNet could effectively predict FTC, provide explanations consistent with pathological knowledge, and improve clinical understanding of the disease.

A Weakly Supervised Deep Learning Framework for Whole Slide Classification to Facilitate Digital Pathology in Animal Study

The pathology of animal studies is crucial for toxicity evaluations and regulatory assessments, but the manual examination of slides by pathologists remains time-consuming and requires extensive training. One inherent challenge in this process is the interobserver variability, which can compromise the consistency and accuracy of a study. Artificial intelligence (AI) has demonstrated its ability to automate similar examinations in clinical applications with enhanced efficiency, consistency, and accuracy. However, training AI models typically relies on costly pixel-level annotation of injured regions and is often not available for animal pathology. To address this, we developed the PathologAI system, a "weakly" supervised approach for WSI classification in rat images without explicit lesion annotation at the pixel level. Using rat liver imaging data from the Open TG-GATEs system, PathologAI was applied to predict necrosis of n = 816 WSIs (377 controls). TG-GATEs studied 170 compounds at three dose levels (low, middle, and high) for four time points (3, 7, 14, and 28 days). PathologAI first preprocessed WSIs at the tile level to generate a high-level representation with a Generative Adversarial Network architecture. The prediction of liver necrosis relied on an ensemble model of 5 CNN classifiers trained on 335 WSIs. The ensemble model achieved notable classification accuracy on the holdout test set: 87% among 87 control slides free of findings, 83% among 120 controls with spontaneous necrosis, 67% among 147 treated animals with spontaneous minimal or slight necrosis, and 59% among 127 treated animals with minimal or slight necrosis caused by the treatment. Importantly, PathologAI was able to discriminate WSIs with spontaneous necrosis from those with treatment related necrosis and discriminated mild lesion level findings (slight vs minimal) and between treatment dose levels. PathologAI could provide an inexpensive and rapid screening tool to assist the digital pathology analysis in preclinical applications and general toxicological studies.

Charting the Next Road Map for CSF Biomarkers in Alzheimer's Disease and Related Dementias

Clinical prediction of underlying pathologic substrates in people with Alzheimer's disease (AD) dementia or related dementia syndromes (ADRD) has limited accuracy. Etiologic biomarkers - including cerebrospinal fluid (CSF) levels of AD proteins and cerebral amyloid PET imaging - have greatly modernized disease-modifying clinical trials in AD, but their integration into medical practice has been slow. Beyond core CSF AD biomarkers (including beta-amyloid 1-42, total tau, and tau phosphorylated at threonine 181), novel biomarkers have been interrogated in single- and multi-centered studies with uneven rigor. Here, we review early expectations for ideal AD/ADRD biomarkers, assess these goals' future applicability, and propose study designs and performance thresholds for meeting these ideals with a focus on CSF biomarkers. We further propose three new characteristics: equity (oversampling of diverse populations in the design and testing of biomarkers), access (reasonable availability to 80% of people at risk for disease, along with pre- and post-biomarker processes), and reliability (thorough evaluation of pre-analytical and analytical factors influencing measurements and performance). Finally, we urge biomarker scientists to balance the desire and evidence for a biomarker to reflect its namesake function, indulge data- as well as theory-driven associations, re-visit the subset of rigorously measured CSF biomarkers in large datasets (such as Alzheimer's disease neuroimaging initiative), and resist the temptation to favor ease over fail-safe in the development phase. This shift from discovery to application, and from suspended disbelief to cogent ingenuity, should allow the AD/ADRD biomarker field to live up to its billing during the next phase of neurodegenerative disease research.

Artificial Intelligence in Breast Imaging: Challenges of Integration Into Clinical Practice

Artificial intelligence (AI) in breast imaging is a rapidly developing field with promising results. Despite the large number of recent publications in this field, unanswered questions have led to limited implementation of AI into daily clinical practice for breast radiologists. This paper provides an overview of the key limitations of AI in breast imaging including, but not limited to, limited numbers of FDA-approved algorithms and annotated data sets with histologic ground truth; concerns surrounding data privacy, security, algorithm transparency, and bias; and ethical issues. Ultimately, the successful implementation of AI into clinical care will require thoughtful action to address these challenges, transparency, and sharing of AI implementation workflows, limitations, and performance metrics within the breast imaging community and other end-users.

Comparing Detection Schemes for Adversarial Images against Deep Learning Models for Cancer Imaging

Deep learning (DL) models have demonstrated state-of-the-art performance in the classification of diagnostic imaging in oncology. However, DL models for medical images can be compromised by adversarial images, where pixel values of input images are manipulated to deceive the DL model. To address this limitation, our study investigates the detectability of adversarial images in oncology using multiple detection schemes. Experiments were conducted on thoracic computed tomography (CT) scans, mammography, and brain magnetic resonance imaging (MRI). For each dataset we trained a convolutional neural network to classify the presence or absence of malignancy. We trained five DL and machine learning (ML)-based detection models and tested their performance in detecting adversarial images. Adversarial images generated using projected gradient descent (PGD) with a perturbation size of 0.004 were detected by the ResNet detection model with an accuracy of 100% for CT, 100% for mammogram, and 90.0% for MRI. Overall, adversarial images were detected with high accuracy in settings where adversarial perturbation was above set thresholds. Adversarial detection should be considered alongside adversarial training as a defense technique to protect DL models for cancer imaging classification from the threat of adversarial images.