Effects of vitamin D supplementation on a deep learning-based mammographic evaluation in SWOG S0812.

Deep learning-based mammographic evaluations could noninvasively assess response to breast cancer chemoprevention. We evaluated change in a convolutional neural network-based breast cancer risk model applied to mammograms among women enrolled in SWOG S0812, which randomly assigned 208 premenopausal high-risk women to receive oral vitamin D3 20 000 IU weekly or placebo for 12 months. We applied the convolutional neural network model to mammograms collected at baseline (n = 109), 12 months (n = 97), and 24 months (n = 67) and compared changes in convolutional neural network-based risk score between treatment groups. Change in convolutional neural network-based risk score was not statistically significantly different between vitamin D and placebo groups at 12 months (0.005 vs 0.002, P = .875) or at 24 months (0.020 vs 0.001, P = .563). The findings are consistent with the primary analysis of S0812, which did not demonstrate statistically significant changes in mammographic density with vitamin D supplementation compared with placebo. There is an ongoing need to evaluate biomarkers of response to novel breast cancer chemopreventive agents.

A Comprehensive Assessment of Radial Scars on Core Needle Biopsy in Patients with or without Breast Cancer: Upgrade Rate and Implications on Management.

BACKGROUND: Radial scars/radial sclerosing lesions (RS) are benign breast lesions identified on core needle biopsy (CNB) which can upgrade to malignancy at excision. There is limited data on RS detection and upgrade rates with more sensitive imaging such as magnetic resonance imaging (MRI) and none during their detection for breast cancer workup and its implication on patient treatment decisions. METHODS: A retrospective institutional study of RS diagnosed on CNB between January 2008 and December 2017 was conducted. Clinicopathologic and radiologic features of RS, patient treatment decisions, upgrade rates and long-term follow-up were examined. RESULTS: We identified 133 patients with RS on CNB, of whom 106 opted for surgery for an upgrade rate to malignancy of 1.9%, 2 patients. Radial scar was diagnosed on mammogram in 60%, MRI in 25% and ultrasound in 15% of patients. In this cohort, 32 patients had their RS detected during breast cancer workup (coexistent group) and they were more likely to have their radial scar detected by MRI (60% vs. 14%, P < .001) and undergo more extensive surgery (94% vs. 75%, P = .02). Among the 27 patients electing observation of their RS, only one (3.7%) developed breast cancer. CONCLUSIONS: Our results show an extremely low upgrade rate to malignancy of RS, regardless if there is coexisting breast cancer elsewhere. Despite this, RS still prompted more extensive surgical excisions. The findings do not support excision of RS even among breast cancer patients when identified at a separate site from their cancer.

Breast Multiparametric MRI for Prediction of Neoadjuvant Chemotherapy Response in Breast Cancer: The BMMR2 Challenge.

Purpose To describe the design, conduct, and results of the Breast Multiparametric MRI for prediction of neoadjuvant chemotherapy Response (BMMR2) challenge. Materials and Methods The BMMR2 computational challenge opened on May 28, 2021, and closed on December 21, 2021. The goal of the challenge was to identify image-based markers derived from multiparametric breast MRI, including diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) MRI, along with clinical data for predicting pathologic complete response (pCR) following neoadjuvant treatment. Data included 573 breast MRI studies from 191 women (mean age [±SD], 48.9 years ± 10.56) in the I-SPY 2/American College of Radiology Imaging Network (ACRIN) 6698 trial (ClinicalTrials.gov: NCT01042379). The challenge cohort was split into training (60%) and test (40%) sets, with teams blinded to test set pCR outcomes. Prediction performance was evaluated by area under the receiver operating characteristic curve (AUC) and compared with the benchmark established from the ACRIN 6698 primary analysis. Results Eight teams submitted final predictions. Entries from three teams had point estimators of AUC that were higher than the benchmark performance (AUC, 0.782 [95% CI: 0.670, 0.893], with AUCs of 0.803 [95% CI: 0.702, 0.904], 0.838 [95% CI: 0.748, 0.928], and 0.840 [95% CI: 0.748, 0.932]). A variety of approaches were used, ranging from extraction of individual features to deep learning and artificial intelligence methods, incorporating DCE and DWI alone or in combination. Conclusion The BMMR2 challenge identified several models with high predictive performance, which may further expand the value of multiparametric breast MRI as an early marker of treatment response. Clinical trial registration no. NCT01042379 Keywords: MRI, Breast, Tumor Response Supplemental material is available for this article. © RSNA, 2024.

Deep Learning Approaches with Digital Mammography for Evaluating Breast Cancer Risk, a Narrative Review.

Breast cancer remains the leading cause of cancer-related deaths in women worldwide. Current screening regimens and clinical breast cancer risk assessment models use risk factors such as demographics and patient history to guide policy and assess risk. Applications of artificial intelligence methods (AI) such as deep learning (DL) and convolutional neural networks (CNNs) to evaluate individual patient information and imaging showed promise as personalized risk models. We reviewed the current literature for studies related to deep learning and convolutional neural networks with digital mammography for assessing breast cancer risk. We discussed the literature and examined the ongoing and future applications of deep learning techniques in breast cancer risk modeling.

Emerging uses of artificial intelligence in breast and axillary ultrasound.

Breast ultrasound is a valuable adjunctive tool to mammography in detecting breast cancer, especially in women with dense breasts. Ultrasound also plays an important role in staging breast cancer by assessing axillary lymph nodes. However, its utility is limited by operator dependence, high recall rate, low positive predictive value and low specificity. These limitations present an opportunity for artificial intelligence (AI) to improve diagnostic performance and pioneer novel uses of ultrasound. Research in developing AI for radiology has flourished over the past few years. A subset of AI, deep learning, uses interconnected computational nodes to form a neural network, which extracts complex visual features from image data to train itself into a predictive model. This review summarizes several key studies evaluating AI programs' performance in predicting breast cancer and demonstrates that AI can assist radiologists and address limitations of ultrasound by acting as a decision support tool. This review also touches on how AI programs allow for novel predictive uses of ultrasound, particularly predicting molecular subtypes of breast cancer and response to neoadjuvant chemotherapy, which have the potential to change how breast cancer is managed by providing non-invasive prognostic and treatment data from ultrasound images. Lastly, this review explores how AI programs demonstrate improved diagnostic accuracy in predicting axillary lymph node metastasis. The limitations and future challenges in developing and implementing AI for breast and axillary ultrasound will also be discussed.

Breast cancer risk prediction combining a convolutional neural network-based mammographic evaluation with clinical factors.

PURPOSE: Deep learning techniques, including convolutional neural networks (CNN), have the potential to improve breast cancer risk prediction compared to traditional risk models. We assessed whether combining a CNN-based mammographic evaluation with clinical factors in the Breast Cancer Surveillance Consortium (BCSC) model improved risk prediction. METHODS: We conducted a retrospective cohort study among 23,467 women, age 35-74, undergoing screening mammography (2014-2018). We extracted electronic health record (EHR) data on risk factors. We identified 121 women who subsequently developed invasive breast cancer at least 1 year after the baseline mammogram. Mammograms were analyzed with a pixel-wise mammographic evaluation using CNN architecture. We used logistic regression models with breast cancer incidence as the outcome and predictors including clinical factors only (BCSC model) or combined with CNN risk score (hybrid model). We compared model prediction performance via area under the receiver operating characteristics curves (AUCs). RESULTS: Mean age was 55.9 years (SD, 9.5) with 9.3% non-Hispanic Black and 36% Hispanic. Our hybrid model did not significantly improve risk prediction compared to the BCSC model (AUC of 0.654 vs 0.624, respectively, p = 0.063). In subgroup analyses, the hybrid model outperformed the BCSC model among non-Hispanic Blacks (AUC 0.845 vs. 0.589; p = 0.026) and Hispanics (AUC 0.650 vs 0.595; p = 0.049). CONCLUSION: We aimed to develop an efficient breast cancer risk assessment method using CNN risk score and clinical factors from the EHR. With future validation in a larger cohort, our CNN model combined with clinical factors may help predict breast cancer risk in a cohort of racially/ethnically diverse women undergoing screening.

Advances in PET/CT Imaging for Breast Cancer Patients and Beyond.

Breast cancer is the most common cancer in women around the world and the fifth leading cause of cancer-related death [...].

Optical coherence tomography holds promise to transform the diagnostic anatomic pathology gross evaluation process.

SIGNIFICANCE: Real-time histology can close a variety of gaps in tissue diagnostics. Currently, gross pathology analysis of excised tissue is dependent upon visual inspection and palpation to identify regions of interest for histopathological processing. Such analysis is limited by the variable correlation between macroscopic and microscopic findings. The current standard of care is costly, burdensome, and inefficient. AIM: We are the first to address this gap by introducing optical coherence tomography (OCT) to be integrated in real-time during the pathology grossing process. APPROACH: This is achieved by our high-resolution, ultrahigh-speed, large field-of-view OCT device designed for this clinical application. RESULTS: We demonstrate the feasibility of imaging tissue sections from multiple human organs (breast, prostate, lung, and pancreas) in a clinical gross pathology setting without interrupting standard workflows. CONCLUSIONS: OCT-based real-time histology evaluation holds promise for addressing a gap that has been present for >100 years.

Use of a convolutional neural network-based mammographic evaluation to predict breast cancer recurrence among women with hormone receptor-positive operable breast cancer.

PURPOSE: We evaluated whether a novel, fully automated convolutional neural network (CNN)-based mammographic evaluation can predict breast cancer relapse among women with operable hormone receptor (HR)-positive breast cancer. METHODS: We conducted a retrospective cohort study among women with stage I-III, HR-positive unilateral breast cancer diagnosed at Columbia University Medical Center from 2007 to 2017, who received adjuvant endocrine therapy and had at least two mammograms (baseline, annual follow-up) of the contralateral unaffected breast for CNN analysis. We extracted demographics, clinicopathologic characteristics, breast cancer treatments, and relapse status from the electronic health record. Our primary endpoint was change in CNN risk score (range, 0-1). We used two-sample t-tests to assess for difference in mean CNN scores between patients who relapsed vs. remained in remission, and conducted Cox regression analyses to assess for association between change in CNN score and breast cancer-free interval (BCFI), adjusting for known prognostic factors. RESULTS: Among 848 women followed for a median of 59 months, there were 67 (7.9%) breast cancer relapses (36 distant, 25 local, 6 new primaries). There was a significant difference in mean absolute change in CNN risk score from baseline to 1-year follow-up between those who relapsed vs. remained in remission (0.001 vs. - 0.022, p = 0.030). After adjustment for prognostic factors, a 0.01 absolute increase in CNN score at 1-year was significantly associated with BCFI, hazard ratio = 1.05 (95% Confidence Interval 1.01-1.09, p = 0.011). CONCLUSION: Short-term change in the CNN-based breast cancer risk model on adjuvant endocrine therapy predicts breast cancer relapse, and warrants further evaluation in prospective studies.

A review of artificial intelligence in mammography.

Breast cancer is the most common cancer among women worldwide. Mammography is the most widely used modality to detect breast cancer. Over the past decade, computer aided detection (CAD) powered by artificial intelligence (AI)/deep learning has shown significant increase in accuracy compared to the traditional CAD. In this review, we aim to summarize the latest developments in the field of AI and mammography and discuss where future progress may lie.

Deep learning prediction of axillary lymph node status using ultrasound images.

OBJECTIVE: To investigate the ability of our convolutional neural network (CNN) to predict axillary lymph node metastasis using primary breast cancer ultrasound (US) images. METHODS: In this IRB-approved study, 338 US images (two orthogonal images) from 169 patients from 1/2014-12/2016 were used. Suspicious lymph nodes were seen on US and patients subsequently underwent core-biopsy. 64 patients had metastatic lymph nodes. A custom CNN was utilized on 248 US images from 124 patients in the training dataset and tested on 90 US images from 45 patients. The CNN was implemented entirely of 3 × 3 convolutional kernels and linear layers. The 9 convolutional kernels consisted of 6 residual layers, totaling 12 convolutional layers. Feature maps were down-sampled using strided convolutions. Dropout with a 0.5 keep probability and L2 normalization was utilized. Training was implemented by using the Adam optimizer and a final SoftMax score threshold of 0.5 from the average of raw logits from each pixel was used for two class classification (metastasis or not). RESULTS: Our CNN achieved an AUC of 0.72 (SD ± 0.08) in predicting axillary lymph node metastasis from US images in the testing dataset. The model had an accuracy of 72.6% (SD ± 8.4) with a sensitivity and specificity of 65.5% (SD ± 28.6) and 78.9% (SD ± 15.1) respectively. Our algorithm is available to be shared for research use. (https://github.com/stmutasa/MetUS). CONCLUSION: It's feasible to predict axillary lymph node metastasis from US images using a deep learning technique. This can potentially aid nodal staging in patients with breast cancer.

The progesterone-receptor modulator, ulipristal acetate, drastically lowers breast cell proliferation.

PURPOSE: The proliferation of breast epithelial cells increases during the luteal phase of the menstrual cycle, when they are exposed to progesterone, suggesting that ulipristal acetate, a selective progestin-receptor modulator (SPRM), may reduce breast cell proliferation with potential use in breast cancer chemoprevention. METHODS: Women aged 18-39 were randomized 1:1 to ulipristal 10-mg daily or to a combination oral contraceptive (COC) for 84 days. Participants underwent a breast biopsy and breast MRI at baseline and at end of study treatment. Proliferation of breast TDLU cells was evaluated by Ki67 immunohistochemical stain. We evaluated the breast MRIs for background parenchymal enhancement (BPE). All slides and images were masked for outcome evaluation. RESULTS: Twenty-eight treatment-compliant participants completed the study; 25 of whom had evaluable Ki67 results at baseline and on-treatment. From baseline to end of treatment, Ki67 % positivity (Ki67%+) decreased a median of 84% in the ulipristal group (N = 13; 2-sided p (2p) = 0.040) versus a median increase of 8% in the COC group (N = 12; 2p = 0.85). Median BPE scores decreased from 3 to 1 in the ulipristal group (p = 0.008) and did not decrease in the COC group. CONCLUSION: Ulipristal was associated with a major decrease in Ki67%+ and BPE. Ulipristal would warrant further investigation for breast cancer chemoprevention were it not for concerns about its liver toxicity. Novel SPRMs without liver toxicity could provide a new approach to breast cancer chemoprevention. TRIAL REGISTRATION: NCT02922127, 4 October 2016.

Weakly Supervised Deep Learning Approach to Breast MRI Assessment.

RATIONALE AND OBJECTIVES: To evaluate a weakly supervised deep learning approach to breast Magnetic Resonance Imaging (MRI) assessment without pixel level segmentation in order to improve the specificity of breast MRI lesion classification. MATERIALS AND METHODS: In this IRB approved study, the dataset consisted of 278,685 image slices from 438 patients. The weakly supervised network was based on the Resnet-101 architecture. Training was implemented using the Adam optimizer and a final SoftMax score threshold of 0.5 was used for two class classification (malignant or benign). 278,685 image slices were combined into 92,895 3-channel images. 79,871 (85%) images were used for training and validation while 13,024 (15%) images were separated for testing. Of the testing dataset, 11,498 (88%) were benign and 1531 (12%) were malignant. Model performance was assessed. RESULTS: The weakly supervised network achieved an AUC of 0.92 (SD ± 0.03) in distinguishing malignant from benign images. The model had an accuracy of 94.2% (SD ± 3.4) with a sensitivity and specificity of 74.4% (SD ± 8.5) and 95.3% (SD ± 3.3) respectively. CONCLUSION: It is feasible to use a weakly supervised deep learning approach to assess breast MRI images without the need for pixel-by-pixel segmentation yielding a high degree of specificity in lesion classification.

Distinguishing benign and malignant lesions on contrast-enhanced breast cone-beam CT with deep learning neural architecture search.

PURPOSE: To utilize a neural architecture search (NAS) approach to develop a convolutional neural network (CNN) method for distinguishing benign and malignant lesions on breast cone-beam CT (BCBCT). METHOD: 165 patients with 114 malignant and 86 benign lesions were collected by two institutions from May 2012 to August 2014. The NAS method autonomously generated a CNN model using one institution's dataset for training (patients/lesions: 71/91) and validation (patients/lesions: 20/23). The model was externally tested on another institution's dataset (patients/lesions: 74/87), and its performance was compared with fine-tuned ResNet-50 models and two breast radiologists who independently read the lesions in the testing dataset without knowing lesion diagnosis. RESULTS: The lesion diameters (mean ± SD) were 18.8 ± 12.9 mm, 22.7 ± 10.5 mm, and 20.0 ± 11.8 mm in the training, validation, and external testing set, respectively. Compared to the best ResNet-50 model, the NAS-generated CNN model performed three times faster and, in the external testing set, achieved a higher (though not statistically different) AUC, with sensitivity (95% CI) and specificity (95% CI) of 0.727, 80% (66-90%), and 60% (42-75%), respectively. Meanwhile, the performances of the NAS-generated CNN and the two radiologists' visual ratings were not statistically different. CONCLUSIONS: Our preliminary results demonstrated that a CNN autonomously generated by NAS performed comparably to pre-trained ResNet models and radiologists in predicting malignant breast lesions on contrast-enhanced BCBCT. In comparison to ResNet, which must be designed by an expert, the NAS approach may be used to automatically generate a deep learning architecture for medical image analysis.

Dynamic Changes of Convolutional Neural Network-based Mammographic Breast Cancer Risk Score Among Women Undergoing Chemoprevention Treatment.

INTRODUCTION: We investigated whether our convolutional neural network (CNN)-based breast cancer risk model is modifiable by testing it on women who had undergone risk-reducing chemoprevention treatment. MATERIALS AND METHODS: We conducted a retrospective cohort study of patients diagnosed with atypical hyperplasia, lobular carcinoma in situ, or ductal carcinoma in situ at our institution from 2007 to 2015. The clinical characteristics, chemoprevention use, and mammography images were extracted from the electronic health records. We classified two groups according to chemoprevention use. Mammograms were performed at baseline and subsequent follow-up evaluations for input to our CNN risk model. The 2 chemoprevention groups were compared for the risk score change from baseline to follow-up. The change categories included stayed high risk, stayed low risk, increased from low to high risk, and decreased from high to low risk. Unordered polytomous regression models were used for statistical analysis, with P < .05 considered statistically significant. RESULTS: Of 541 patients, 184 (34%) had undergone chemoprevention treatment (group 1) and 357 (66%) had not (group 2). Using our CNN breast cancer risk score, significantly more women in group 1 had shown a decrease in breast cancer risk compared with group 2 (33.7% vs. 22.9%; P < .01). Significantly fewer women in group 1 had an increase in breast cancer risk compared with group 2 (11.4% vs. 20.2%; P < .01). On multivariate analysis, an increase in breast cancer risk predicted by our model correlated negatively with the use of chemoprevention treatment (P = .02). CONCLUSIONS: Our CNN-based breast cancer risk score is modifiable with potential utility in assessing the efficacy of known chemoprevention agents and testing new chemoprevention strategies.

Harmonic motion imaging of human breast masses: an in vivo clinical feasibility.

Non-invasive diagnosis of breast cancer is still challenging due to the low specificity of the imaging modalities that calls for unnecessary biopsies. The diagnostic accuracy can be improved by assessing the breast tissue mechanical properties associated with pathological changes. Harmonic motion imaging (HMI) is an elasticity imaging technique that uses acoustic radiation force to evaluate the localized mechanical properties of the underlying tissue. Herein, we studied the in vivo feasibility of a clinical HMI system to differentiate breast tumors based on their relative HMI displacements, in human subjects. We performed HMI scans in 10 female subjects with breast masses: five benign and five malignant masses. Results revealed that both benign and malignant masses were stiffer than the surrounding tissues. However, malignant tumors underwent lower mean HMI displacement (1.1 ± 0.5 µm) compared to benign tumors (3.6 ± 1.5 µm) and the adjacent non-cancerous tissue (6.4 ± 2.5 µm), which allowed to differentiate between tumor types. Additionally, the excised breast specimens of the same patients (n = 5) were imaged post-surgically, where there was an excellent agreement between the in vivo and ex vivo findings, confirmed with histology. Higher displacement contrast between cancerous and non-cancerous tissue was found ex vivo, potentially due to the lower nonlinearity in the elastic properties of ex vivo tissue. This preliminary study lays the foundation for the potential complementary application of HMI in clinical practice in conjunction with the B-mode to classify suspicious breast masses.

A novel CNN algorithm for pathological complete response prediction using an I-SPY TRIAL breast MRI database.

PURPOSE: To apply our convolutional neural network (CNN) algorithm to predict neoadjuvant chemotherapy (NAC) response using the I-SPY TRIAL breast MRI dataset. METHODS: From the I-SPY TRIAL breast MRI database, 131 patients from 9 institutions were successfully downloaded for analysis. First post-contrast MRI images were used for 3D segmentation using 3D slicer. Our CNN was implemented entirely of 3 × 3 convolutional kernels and linear layers. The convolutional kernels consisted of 6 residual layers, totaling 12 convolutional layers. Dropout with a 0.5 keep probability and L2 normalization was utilized. Training was implemented by using the Adam optimizer. A 5-fold cross validation was used for performance evaluation. Software code was written in Python using the TensorFlow module on a Linux workstation with one NVidia Titan X GPU. RESULTS: Of 131 patients, 40 patients achieved pCR following NAC (group 1) and 91 patients did not achieve pCR following NAC (group 2). Diagnostic accuracy of our CNN two classification model distinguishing patients with pCR vs non-pCR was 72.5 (SD ± 8.4), with sensitivity 65.5% (SD ± 28.1) and specificity of 78.9% (SD ± 15.2). The area under a ROC Curve (AUC) was 0.72 (SD ± 0.08). CONCLUSION: It is feasible to use our CNN algorithm to predict NAC response in patients using a multi-institution dataset.

Prospective Analysis Using a Novel CNN Algorithm to Distinguish Atypical Ductal Hyperplasia From Ductal Carcinoma in Situ in Breast.

INTRODUCTION: We previously developed a convolutional neural networks (CNN)-based algorithm to distinguish atypical ductal hyperplasia (ADH) from ductal carcinoma in situ (DCIS) using a mammographic dataset. The purpose of this study is to further validate our CNN algorithm by prospectively analyzing an unseen new dataset to evaluate the diagnostic performance of our algorithm. MATERIALS AND METHODS: In this institutional review board-approved study, a new dataset composed of 280 unique mammographic images from 140 patients was used to test our CNN algorithm. All patients underwent stereotactic-guided biopsy of calcifications and underwent surgical excision with available final pathology. The ADH group consisted of 122 images from 61 patients with the highest pathology diagnosis of ADH. The DCIS group consisted of 158 images from 79 patients with the highest pathology diagnosis of DCIS. Two standard mammographic magnification views (craniocaudal and mediolateral/lateromedial) of the calcifications were used for analysis. Calcifications were segmented using an open source software platform 3D slicer and resized to fit a 128 × 128 pixel bounding box. Our previously developed CNN algorithm was used. Briefly, a 15 hidden layer topology was used. The network architecture contained 5 residual layers and dropout of 0.25 after each convolution. Diagnostic performance metrics were analyzed including sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve. The "positive class" was defined as the pure ADH group in this study and thus specificity represents minimizing the amount of falsely labeled pure ADH cases. RESULTS: Area under the receiver operating characteristic curve was 0.90 (95% confidence interval, ± 0.04). Diagnostic accuracy, sensitivity, and specificity was 80.7%, 63.9%, and 93.7%, respectively. CONCLUSION: Prospectively tested on new unseen data, our CNN algorithm distinguished pure ADH from DCIS using mammographic images with high specificity.

Channel width optimized neural networks for liver and vessel segmentation in liver iron quantification.

INTRODUCTION: MRI T2* relaxometry protocols are often used for Liver Iron Quantification in patients with hemochromatosis. Several methods exist to semi-automatically segment parenchyma and exclude vessels for this calculation. PURPOSE: To determine if inclusion of multiple echoes inputs to Convolutional Neural Networks (CNN) improves automated liver and vessel segmentation in MRI T2* relaxometry protocols and to determine if the resultant segmentations agree with manual segmentations for liver iron quantification analysis. METHODS: Multi echo Gradient Recalled Echo (GRE) MRI sequence for T2* relaxometry was performed for 79 exams on 31 patients with hemochromatosis for iron quantification analysis. 275 axial liver slices were manually segmented as ground truth masks. A batch normalized U-Net with variable input width to incorporate multiple echoes is used for segmentation, using DICE as the accuracy metric. ANOVA is used to evaluate significance of channel width changes in segmentation accuracy. Linear regression is used to model the relationship of channel width on segmentation accuracy. Liver segmentations are applied to relaxometry data to calculate liver T2* yielding liver iron concentration(LIC) derived from literature based calibration curves. Manual and CNN based LIC values are compared with Pearson correlation. Bland altman plots are used to visualize differences between manual and CNN based LIC values. RESULTS: Performance metrics are tested on 55 hold out slices. Linear regression indicates that there is a monotonic increase of DICE with increasing channel depth (p = 0.001) with a slope of 3.61e-3. ANOVA indicates a significant increase segmentation accuracy over single channel starting at 3 channels. Incorporation of all channels results in an average DICE of 0.86, an average increase of 0.07 over single channel. The calculated LIC from CNN segmented livers agrees well with manual segmentation (R = 0.998, slope = 0.914, p«0.001), with an average absolute difference 0.27 ± 0.99 mg Fe/g or 1.34 ± 4.3%. CONCLUSION: More input echoes yields higher model accuracy until the noise floor. Echos beyond the first three echo times in GRE based T2* relaxometry do not contribute significant information for segmentation of liver for LIC calculation. Deep learning models with three channel width allow for generalization of model to protocols of more than three echoes, effectively a universal requirement for relaxometry. Deep learning segmentations achieve a good accuracy compared with manual segmentations with minimal preprocessing. Liver iron values calculated from hand segmented liver and Neural network segmented liver were not statistically different from each other.

Should We Ignore, Follow, or Biopsy? Impact of Artificial Intelligence Decision Support on Breast Ultrasound Lesion Assessment.

OBJECTIVE. The objective of this study was to assess the impact of artificial intelligence (AI)-based decision support (DS) on breast ultrasound (US) lesion assessment. MATERIALS AND METHODS. A multicenter retrospective review of 900 breast lesions (470/900 [52.2%] benign; 430/900 [47.8%] malignant) on US by 15 physicians (11 radiologists, two surgeons, two obstetrician/gynecologists). An AI system (Koios DS for Breast, Koios Medical) evaluated images and assigned them to one of four categories: benign, probably benign, suspicious, and probably malignant. Each reader reviewed cases twice: 750 cases with US only or with US plus DS; 4 weeks later, cases were reviewed in the opposite format. One hundred fifty additional cases were presented identically in each session. DS and reader sensitivity, specificity, and positive likelihood ratios (PLRs) were calculated as well as reader AUCs with and without DS. The Kendall τ-b correlation coefficient was used to assess intraand interreader variability. RESULTS. Mean reader AUC for cases reviewed with US only was 0.83 (95% CI, 0.78-0.89); for cases reviewed with US plus DS, mean AUC was 0.87 (95% CI, 0.84-0.90). PLR for the DS system was 1.98 (95% CI, 1.78-2.18) and was higher than the PLR for all readers but one. Fourteen readers had better AUC with US plus DS than with US only. Mean Kendall τ-b for US-only interreader variability was 0.54 (95% CI, 0.53-0.55); for US plus DS, it was 0.68 (95% CI, 0.67-0.69). Intrareader variability improved with DS; class switching (defined as crossing from BI-RADS category 3 to BI-RADS category 4A or above) occurred in 13.6% of cases with US only versus 10.8% of cases with US plus DS (p = 0.04). CONCLUSION. AI-based DS improves accuracy of sonographic breast lesion assessment while reducing inter- and intraobserver variability.