AI analytics can be used as imaging biomarkers for predicting invasive upgrade of ductal carcinoma in situ.

OBJECTIVES: To evaluate whether the quantitative abnormality scores provided by artificial intelligence (AI)-based computer-aided detection/diagnosis (CAD) for mammography interpretation can be used to predict invasive upgrade in ductal carcinoma in situ (DCIS) diagnosed on percutaneous biopsy. METHODS: Four hundred forty DCIS in 420 women (mean age, 52.8 years) diagnosed via percutaneous biopsy from January 2015 to December 2019 were included. Mammographic characteristics were assessed based on imaging features (mammographically occult, mass/asymmetry/distortion, calcifications only, and combined mass/asymmetry/distortion with calcifications) and BI-RADS assessments. Routine pre-biopsy 4-view digital mammograms were analyzed using AI-CAD to obtain abnormality scores (AI-CAD score, ranging 0-100%). Multivariable logistic regression was performed to identify independent predictive mammographic variables after adjusting for clinicopathological variables. A subgroup analysis was performed with mammographically detected DCIS. RESULTS: Of the 440 DCIS, 117 (26.6%) were upgraded to invasive cancer. Three hundred forty-one (77.5%) DCIS were detected on mammography. The multivariable analysis showed that combined features (odds ratio (OR): 2.225, p = 0.033), BI-RADS 4c or 5 assessments (OR: 2.473, p = 0.023 and OR: 5.190, p < 0.001, respectively), higher AI-CAD score (OR: 1.009, p = 0.007), AI-CAD score ≥ 50% (OR: 1.960, p = 0.017), and AI-CAD score ≥ 75% (OR: 2.306, p = 0.009) were independent predictors of invasive upgrade. In mammographically detected DCIS, combined features (OR: 2.194, p = 0.035), and higher AI-CAD score (OR: 1.008, p = 0.047) were significant predictors of invasive upgrade. CONCLUSION: The AI-CAD score was an independent predictor of invasive upgrade for DCIS. Higher AI-CAD scores, especially in the highest quartile of ≥ 75%, can be used as an objective imaging biomarker to predict invasive upgrade in DCIS diagnosed with percutaneous biopsy. CRITICAL RELEVANCE STATEMENT: Noninvasive imaging features including the quantitative results of AI-CAD for mammography interpretation were independent predictors of invasive upgrade in lesions initially diagnosed as ductal carcinoma in situ via percutaneous biopsy and therefore may help decide the direction of surgery before treatment. KEY POINTS: • Predicting ductal carcinoma in situ upgrade is important, yet there is a lack of conclusive non-invasive biomarkers. • AI-CAD scores-raw numbers, ≥ 50%, and ≥ 75%-predicted ductal carcinoma in situ upgrade independently. • Quantitative AI-CAD results may help predict ductal carcinoma in situ upgrade and guide patient management.

Unenhanced Breast MRI With Diffusion-Weighted Imaging for Breast Cancer Detection: Effects of Training on Performance and Agreement of Subspecialty Radiologists.

OBJECTIVE: To investigate whether reader training improves the performance and agreement of radiologists in interpreting unenhanced breast magnetic resonance imaging (MRI) scans using diffusion-weighted imaging (DWI). MATERIALS AND METHODS: A study of 96 breasts (35 cancers, 24 benign, and 37 negative) in 48 asymptomatic women was performed between June 2019 and October 2020. High-resolution DWI with b-values of 0, 800, and 1200 sec/mm² was performed using a 3.0-T system. Sixteen breast radiologists independently reviewed the DWI, apparent diffusion coefficient maps, and T1-weighted MRI scans and recorded the Breast Imaging Reporting and Data System (BI-RADS) category for each breast. After a 2-h training session and a 5-month washout period, they re-evaluated the BI-RADS categories. A BI-RADS category of 4 (lesions with at least two suspicious criteria) or 5 (more than two suspicious criteria) was considered positive. The per-breast diagnostic performance of each reader was compared between the first and second reviews. Inter-reader agreement was evaluated using a multi-rater κ analysis and intraclass correlation coefficient (ICC). RESULTS: Before training, the mean sensitivity, specificity, and accuracy of the 16 readers were 70.7% (95% confidence interval [CI]: 59.4-79.9), 90.8% (95% CI: 85.6-94.2), and 83.5% (95% CI: 78.6-87.4), respectively. After training, significant improvements in specificity (95.2%; 95% CI: 90.8-97.5; P = 0.001) and accuracy (85.9%; 95% CI: 80.9-89.8; P = 0.01) were observed, but no difference in sensitivity (69.8%; 95% CI: 58.1-79.4; P = 0.58) was observed. Regarding inter-reader agreement, the κ values were 0.57 (95% CI: 0.52-0.63) before training and 0.68 (95% CI: 0.62-0.74) after training, with a difference of 0.11 (95% CI: 0.02-0.18; P = 0.01). The ICC was 0.73 (95% CI: 0.69-0.74) before training and 0.79 (95% CI: 0.76-0.80) after training (P = 0.002). CONCLUSION: Brief reader training improved the performance and agreement of interpretations by breast radiologists using unenhanced MRI with DWI.

Screening in Patients With Dense Breasts: Comparison of Mammography, Artificial Intelligence, and Supplementary Ultrasound.

BACKGROUND. Screening mammography has decreased performance in patients with dense breasts. Supplementary screening ultrasound is a recommended option in such patients, although it has yielded mixed results in prior investigations. OBJECTIVE. The purpose of this article is to compare the performance characteristics of screening mammography alone, standalone artificial intelligence (AI), ultrasound alone, and mammography in combination with AI and/or ultrasound in patients with dense breasts. METHODS. This retrospective study included 1325 women (mean age, 53 years) with dense breasts who underwent both screening mammography and supplementary breast ultrasound within a 1-month interval from January 2017 to December 2017; prior mammography and prior ultrasound examinations were available for comparison in 91.2% and 91.8%, respectively. Mammography and ultrasound examinations were interpreted by one of 15 radiologists (five staff; 10 fellows); clinical reports were used for the present analysis. A commercial AI tool was used to retrospectively evaluate mammographic examinations for presence of cancer. Screening performances were compared among mammography, AI, ultrasound, and test combinations, using generalized estimating equations. Benign diagnoses required 24 months or longer of imaging stability. RESULTS. Twelve cancers (six invasive ductal carcinoma; six ductal carcinoma in situ) were diagnosed. Mammography, standalone AI, and ultrasound showed cancer detection rates (per 1000 patients) of 6.0, 6.8, and 6.0 (all p > .05); recall rates of 4.4%, 11.9%, and 9.2% (all p < .05); sensitivity of 66.7%, 75.0%, and 66.7% (all p > .05); specificity of 96.2%, 88.7%, and 91.3% (all p < .05); and accuracy of 95.9%, 88.5%, and 91.1% (all p < .05). Mammography with AI, mammography with ultrasound, and mammography with both ultrasound and AI showed cancer detection rates of 7.5, 9.1, and 9.1 (all p > .05); recall rates of 14.9, 11.7, and 21.4 (all p < .05); sensitivity of 83.3%, 100.0%, and 100.0% (all p > .05); specificity of 85.8%, 89.1%, and 79.4% (all p < .05); and accuracy of 85.7%, 89.2%, and 79.5% (all p < .05). CONCLUSION. Mammography with supplementary ultrasound showed higher accuracy, higher specificity, and lower recall rate in comparison with mammography with AI and in comparison with mammography with both ultrasound and AI. CLINICAL IMPACT. The findings fail to show benefit of AI with respect to screening mammography performed with supplementary breast ultrasound in patients with dense breasts.

Immune response and mesenchymal transition of papillary thyroid carcinoma reflected in ultrasonography features assessed by radiologists and deep learning.

INTRODUCTION: Ultrasonography (US) features of papillary thyroid cancers (PTCs) are used to select nodules for biopsy due to their association with tumor behavior. However, the molecular biological mechanisms that lead to the characteristic US features of PTCs are largely unknown. OBJECTIVES: This study aimed to investigate the molecular biological mechanisms behind US features assessed by radiologists and three convolutional neural networks (CNN) through transcriptome analysis. METHODS: Transcriptome data from 273 PTC tissue samples were generated and differentially expressed genes (DEGs) were identified according to US feature. Pathway enrichment analyses were also conducted by gene set enrichment analysis (GSEA) and ClusterProfiler according to assessments made by radiologists and three CNNs - CNN1 (ResNet50), CNN2 (ResNet101) and CNN3 (VGG16). Signature gene scores for PTCs were calculated by single-sample GSEA (ssGSEA). RESULTS: Individual suspicious US features consistently suggested an upregulation of genes related to immune response and epithelial-mesenchymal transition (EMT). Likewise, PTCs assessed as positive by radiologists and three CNNs showed the coordinate enrichment of similar gene sets with abundant immune and stromal components. However, PTCs assessed as positive by radiologists had the highest number of DEGs, and those assessed as positive by CNN3 had more diverse DEGs and gene sets compared to CNN1 or CNN2. The percentage of PTCs assessed as positive or negative concordantly by radiologists and three CNNs was 85.6% (231/273) and 7.1% (3/273), respectively. CONCLUSION: US features assessed by radiologists and CNNs revealed molecular biologic features and tumor microenvironment in PTCs.

International Expert Consensus on US Lexicon for Thyroid Nodules.

Multiple US-based systems for risk stratification of thyroid nodules are in use worldwide. Unfortunately, the malignancy probability assigned to a nodule varies, and terms and definitions are not consistent, leading to confusion and making it challenging to compare study results and craft revisions. Consistent application of these systems is further hampered by interobserver variability in identifying the sonographic features on which they are founded. In 2018, an international multidisciplinary group of 19 physicians with expertise in thyroid sonography (termed the International Thyroid Nodule Ultrasound Working Group) was convened with the goal of developing an international system, tentatively called the International Thyroid Imaging Reporting and Data System, or I-TIRADS, in two phases: (phase I) creation of a lexicon and atlas of US descriptors of thyroid nodules and (phase II) development of a system that estimates the malignancy risk of a thyroid nodule. This article presents the methods and results of phase I. The purpose herein is to show what has been accomplished thus far, as well as generate interest in and support for this effort in the global thyroid community.

Learnability of Thyroid Nodule Assessment on Ultrasonography: Using a Big Data Set.

OBJECTIVE: The aims of the work described here were to evaluate the learnability of thyroid nodule assessment on ultrasonography (US) using a big data set of US images and to evaluate the diagnostic utilities of artificial intelligence computer-aided diagnosis (AI-CAD) used by readers with varying experience to differentiate benign and malignant thyroid nodules. METHODS: Six college freshmen independently studied the "learning set" composed of images of 13,560 thyroid nodules, and their diagnostic performance was evaluated after their daily learning sessions using the "test set" composed of images of 282 thyroid nodules. The diagnostic performance of two residents and an experienced radiologist was evaluated using the same "test set." After an initial diagnosis, all readers once again evaluated the "test set" with the assistance of AI-CAD. RESULTS: Diagnostic performance of almost all students increased after the learning program. Although the mean areas under the receiver operating characteristic curves (AUROCs) of residents and the experienced radiologist were significantly higher than those of students, the AUROCs of five of the six students did not differ significantly compared with that of the one resident. With the assistance of AI-CAD, sensitivity significantly increased in three students, specificity in one student, accuracy in four students and AUROC in four students. Diagnostic performance of the two residents and the experienced radiologist was better with the assistance of AI-CAD. CONCLUSION: A self-learning method using a big data set of US images has potential as an ancillary tool alongside traditional training methods. With the assistance of AI-CAD, the diagnostic performance of readers with varying experience in thyroid imaging could be further improved.

Artificial intelligence-based computer-assisted detection/diagnosis (AI-CAD) for screening mammography: Outcomes of AI-CAD in the mammographic interpretation workflow.

Purpose: To evaluate the stand-alone diagnostic performances of AI-CAD and outcomes of AI-CAD detected abnormalities when applied to the mammographic interpretation workflow. Methods: From January 2016 to December 2017, 6499 screening mammograms of 5228 women were collected from a single screening facility. Historic reads of three radiologists were used as radiologist interpretation. A commercially-available AI-CAD was used for analysis. One radiologist not involved in interpretation had retrospectively reviewed the abnormality features and assessed the significance (negligible vs. need recall) of the AI-CAD marks. Ground truth in terms of cancer, benign or absence of abnormality was confirmed according to histopathologic diagnosis or negative results on the next-round screen. Results: Of the 6499 mammograms, 6282 (96.7%) were in the negative, 189 (2.9%) were in the benign, and 28 (0.4%) were in the cancer group. AI-CAD detected 5 (17.9%, 5 of 28) of the 9 cancers that were intially interpreted as negative. Of the 648 AI-CAD recalls, 89.0% (577 of 648) were marks seen on examinations in the negative group, and 267 (41.2%) of the AI-CAD marks were considered to be negligible. Stand-alone AI-CAD has significantly higher recall rates (10.0% vs. 3.4%, P < 0.001) with comparable sensitivity and cancer detection rates (P = 0.086 and 0.102, respectively) when compared to the radiologists' interpretation. Conclusion: AI-CAD detected 17.9% additional cancers on screening mammography that were initially overlooked by the radiologists. In spite of the additional cancer detection, AI-CAD had significantly higher recall rates in the clinical workflow, in which 89.0% of AI-CAD marks are on negative mammograms.

Diagnosis of thyroid micronodules on ultrasound using a deep convolutional neural network.

To assess the performance of deep convolutional neural network (CNN) to discriminate malignant and benign thyroid nodules < 10 mm in size and compare the diagnostic performance of CNN with those of radiologists. Computer-aided diagnosis was implemented with CNN and trained using ultrasound (US) images of 13,560 nodules ≥ 10 mm in size. Between March 2016 and February 2018, US images of nodules < 10 mm were retrospectively collected at the same institution. All nodules were confirmed as malignant or benign from aspirate cytology or surgical histology. Diagnostic performances of CNN and radiologists were assessed and compared for area under curve (AUC), sensitivity, specificity, accuracy, positive predictive value, and negative predictive value. Subgroup analyses were performed based on nodule size with a cut-off value of 5 mm. Categorization performances of CNN and radiologists were also compared. A total of 370 nodules from 362 consecutive patients were assessed. CNN showed higher negative predictive value (35.3% vs. 22.6%, P = 0.048) and AUC (0.66 vs. 0.57, P = 0.04) than radiologists. CNN also showed better categorization performance than radiologists. In the subgroup of nodules ≤ 5 mm, CNN showed higher AUC (0.63 vs. 0.51, P = 0.08) and specificity (68.2% vs. 9.1%, P < 0.001) than radiologists. Convolutional neural network trained with thyroid nodules ≥ 10 mm in size showed overall better diagnostic performance than radiologists in the diagnosis and categorization of thyroid nodules < 10 mm, especially in nodules ≤ 5 mm.

Changes in Automated Mammographic Breast Density Can Predict Pathological Response After Neoadjuvant Chemotherapy in Breast Cancer.

OBJECTIVE: Mammographic density is an independent risk factor for breast cancer that can change after neoadjuvant chemotherapy (NCT). This study aimed to evaluate percent changes in volumetric breast density (ΔVbd%) before and after NCT measured automatically and determine its value as a predictive marker of pathological response to NCT. MATERIALS AND METHODS: A total of 357 patients with breast cancer treated between January 2014 and December 2016 were included. An automated volumetric breast density (Vbd) measurement method was used to calculate Vbd on mammography before and after NCT. Patients were divided into three groups according to ΔVbd%, calculated as follows: Vbd (post-NCT - pre-NCT)/pre-NCT Vbd × 100 (%). The stable, decreased, and increased groups were defined as -20% ≤ ΔVbd% ≤ 20%, ΔVbd% < -20%, and ΔVbd% > 20%, respectively. Pathological complete response (pCR) was considered to be achieved after NCT if there was no evidence of invasive carcinoma in the breast or metastatic tumors in the axillary and regional lymph nodes on surgical pathology. The association between ΔVbd% grouping and pCR was analyzed using univariable and multivariable logistic regression analyses. RESULTS: The interval between the pre-NCT and post-NCT mammograms ranged from 79 to 250 days (median, 170 days). In the multivariable analysis, ΔVbd% grouping (odds ratio for pCR of 0.420 [95% confidence interval, 0.195-0.905; P = 0.027] for the decreased group compared with the stable group), N stage at diagnosis, histologic grade, and breast cancer subtype were significantly associated with pCR. This tendency was more evident in the luminal B-like and triple-negative subtypes. CONCLUSION: ΔVbd% was associated with pCR in breast cancer after NCT, with the decreased group showing a lower rate of pCR than the stable group. Automated measurement of ΔVbd% may help predict the NCT response and prognosis in breast cancer.

Standalone AI for Breast Cancer Detection at Screening Digital Mammography and Digital Breast Tomosynthesis: A Systematic Review and Meta-Analysis.

Background There is considerable interest in the potential use of artificial intelligence (AI) systems in mammographic screening. However, it is essential to critically evaluate the performance of AI before it can become a modality used for independent mammographic interpretation. Purpose To evaluate the reported standalone performances of AI for interpretation of digital mammography and digital breast tomosynthesis (DBT). Materials and Methods A systematic search was conducted in PubMed, Google Scholar, Embase (Ovid), and Web of Science databases for studies published from January 2017 to June 2022. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) values were reviewed. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and Comparative (QUADAS-2 and QUADAS-C, respectively). A random effects meta-analysis and meta-regression analysis were performed for overall studies and for different study types (reader studies vs historic cohort studies) and imaging techniques (digital mammography vs DBT). Results In total, 16 studies that include 1 108 328 examinations in 497 091 women were analyzed (six reader studies, seven historic cohort studies on digital mammography, and four studies on DBT). Pooled AUCs were significantly higher for standalone AI than radiologists in the six reader studies on digital mammography (0.87 vs 0.81, P = .002), but not for historic cohort studies (0.89 vs 0.96, P = .152). Four studies on DBT showed significantly higher AUCs in AI compared with radiologists (0.90 vs 0.79, P < .001). Higher sensitivity and lower specificity were seen for standalone AI compared with radiologists. Conclusion Standalone AI for screening digital mammography performed as well as or better than radiologists. Compared with digital mammography, there is an insufficient number of studies to assess the performance of AI systems in the interpretation of DBT screening examinations. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Scaranelo in this issue.

Radiomics Analysis of Gray-Scale Ultrasonographic Images of Papillary Thyroid Carcinoma > 1 cm: Potential Biomarker for the Prediction of Lymph Node Metastasis.

Purpose: This study aimed to investigate radiomics analysis of ultrasonographic images to develop a potential biomarker for predicting lymph node metastasis in papillary thyroid carcinoma (PTC) patients. Materials and Methods: This study included 431 PTC patients from August 2013 to May 2014 and classified them into the training and validation sets. A total of 730 radiomics features, including texture matrices of gray-level co-occurrence matrix and gray-level run-length matrix and single-level discrete two-dimensional wavelet transform and other functions, were obtained. The least absolute shrinkage and selection operator method was used for selecting the most predictive features in the training data set. Results: Lymph node metastasis was associated with the radiomics score (p < 0.001). It was also associated with other clinical variables such as young age (p = 0.007) and large tumor size (p = 0.007). The area under the receiver operating characteristic curve was 0.687 (95% confidence interval: 0.616-0.759) for the training set and 0.650 (95% confidence interval: 0.575-0.726) for the validation set. Conclusion: This study showed the potential of ultrasonography-based radiomics to predict cervical lymph node metastasis in patients with PTC; thus, ultrasonography-based radiomics can act as a biomarker for PTC.

Artificial intelligence assistance for women who had spot compression view: reducing recall rates for digital mammography.

BACKGROUND: Mammography yields inevitable recall for indeterminate findings that need to be confirmed with additional views. PURPOSE: To explore whether the artificial intelligence (AI) algorithm for mammography can reduce false-positive recall in patients who undergo the spot compression view. MATERIAL AND METHODS: From January to December 2017, 236 breasts from 225 women who underwent the spot compression view due to focal asymmetry, mass, or architectural distortion on standard digital mammography were included. Three readers who were blinded to the study purpose, patient information, previous mammograms, following spot compression views, and any clinical or pathologic reports retrospectively reviewed 236 standard mammograms and determined the necessity of patient recall and the probability of malignancy per breast, first without and then with AI assistance. The performances of AI and the readers were evaluated with the recall rate, area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy. RESULTS: Among 236 examinations, 8 (3.4%) were cancers and 228 (96.6%) were benign. The recall rates of all three readers significantly decreased with AI assistance (P < 0.05). The reader-averaged recall rates significantly decreased with AI assistance regardless of breast composition (fatty breasts: 32.7% to 24.1%m P = 0.002; dense breasts: 33.6% to 21.2%, P < 0.001). The reader-averaged AUC increased with AI assistance and was comparable to that of standalone AI (0.835 vs. 0.895; P = 0.234). The reader-averaged specificity (71.2% to 79.8%, P < 0.001) and accuracy (71.3% to 79.7%, P < 0.001) significantly improved with AI assistance. CONCLUSION: AI assistance significantly reduced false-positive recall without compromising cancer detection in women with focal asymmetry, mass, or architectural distortion on standard digital mammography regardless of mammographic breast density.

Fine Needle Aspiration Cytology vs. Core Needle Biopsy for Thyroid Nodules: A Prospective, Experimental Study Using Surgical Specimen.

Purpose: To evaluate and compare the diagnostic outcomes of ultrasonography (US)-guided fine needle aspiration (FNA) and core needle biopsy (CNB) performed on the same thyroid nodule using a surgical specimen for direct comparison. Materials and Methods: We included 89 thyroid nodules from 88 patients from February 2015 to January 2016. The inclusion criterion was thyroid nodules measuring ≥ 20 mm (mean size: 40.0 ± 15.3 mm). Immediately after surgical resection, FNA and subsequent CNB were performed on the surgical specimen under US guidance. FNA and CNB cytopathologic results on the specimen were compared with the surgical diagnosis. Results: Among the 89 nodules, 30 were malignant and 59 were benign. Significantly higher inconclusive rates were seen in FNA for malignant than benign nodules (80.0% vs. 39.0%, p < 0.001). For CNB, conclusive and inconclusive rates did not differ between benign and malignant nodules (p = 0.796). Higher inconclusive rates were seen for FNA among cancers regardless of US features, and in the subgroup of size ≥ 40 mm (62.5% vs. 22.9%, p = 0.028). Eleven cancers were diagnosed with CNB (36.7%, 11/30), while none was diagnosed using FNA. Conclusion: In this experimental study using surgical specimens, CNB showed a potential to provide improved diagnostic sensitivity for thyroid cancer, especially when a conclusive diagnosis is limited with FNA.

Use of long non-coding RNAs for the molecular diagnosis of papillary thyroid cancer.

Objective: Improved molecular testing for common somatic mutations and the identification of mRNA and microRNA expression classifiers are promising approaches for the diagnosis of thyroid nodules. However, there is a need to improve the diagnostic accuracy of such tests for identifying thyroid cancer. Recent findings have revealed a crucial role of long non-coding RNAs (lncRNAs) in gene modulation. Thus, we aimed to evaluate the diagnostic value of selected lncRNAs from The Atlas of Noncoding RNAs in Cancer (TANRIC) thyroid cancer dataset. Methods: LncRNAs in TANRIC thyroid cancer dataset that have significantly increased or decreased expression in papillary thyroid cancer (PTC) tissues were selected as candidates for PTC diagnosis. Surgical specimens from patients who underwent thyroidectomy were used to determine the separation capability of candidate lncRNAs between malignant and benign nodules. Fine needle aspiration samples were obtained and screened for candidate lncRNAs to verify their diagnostic value. Results: LRRC52-AS1, LINC02471, LINC02082, UNC5B-AS1, LINC02408, MPPED2-AS1, LNCNEF, LOC642484, ATP6V0E2-AS1, and LOC100129129 were selected as the candidate lncRNAs. LRRC52-AS1, LINC02082, UNC5B-AS1, MPPED2-AS1, LNCNEF, and LOC100129129 expression levels were significantly increased or decreased in malignant nodules compared to those in benign nodules and paired normal thyroid tissues. The combination of LRRC52-AS1, LINC02082, and UNC5B-AS1 showed favorable results for the diagnosis of PTC from fine needle aspirates, with 88.9% sensitivity and 100.0% specificity. Conclusions: LncRNA expression analysis is a promising approach for advancing the molecular diagnosis of PTC. Further studies are needed to identify lncRNAs of additional diagnostic value.

Application of Artificial Intelligence Computer-Assisted Diagnosis Originally Developed for Thyroid Nodules to Breast Lesions on Ultrasound.

As thyroid and breast cancer have several US findings in common, we applied an artificial intelligence computer-assisted diagnosis (AI-CAD) software originally developed for thyroid nodules to breast lesions on ultrasound (US) and evaluated its diagnostic performance. From January 2017 to December 2017, 1042 breast lesions (mean size 20.2 ± 11.8 mm) of 1001 patients (mean age 45.9 ± 12.9 years) who underwent US-guided core-needle biopsy were included. An AI-CAD software that was previously trained and validated with thyroid nodules using the convolutional neural network was applied to breast nodules. There were 665 benign breast lesions (63.0%) and 391 breast cancers (37.0%). The area under the receiver operating characteristic curve (AUROC) of AI-CAD to differentiate breast lesions was 0.678 (95% confidence interval: 0.649, 0.707). After fine-tuning AI-CAD with 1084 separate breast lesions, the diagnostic performance of AI-CAD markedly improved (AUC 0.841). This was significantly higher than that of radiologists when the cutoff category was BI-RADS 4a (AUC 0.621, P < 0.001), but lower when the cutoff category was BI-RADS 4b (AUC 0.908, P < 0.001). When applied to breast lesions, the diagnostic performance of an AI-CAD software that had been developed for differentiating malignant and benign thyroid nodules was not bad. However, an organ-specific approach guarantees better diagnostic performance despite the similar US features of thyroid and breast malignancies.

Differing benefits of artificial intelligence-based computer-aided diagnosis for breast US according to workflow and experience level.

PURPOSE: This study evaluated how artificial intelligence-based computer-assisted diagnosis (AICAD) for breast ultrasonography (US) influences diagnostic performance and agreement between radiologists with varying experience levels in different workflows. METHODS: Images of 492 breast lesions (200 malignant and 292 benign masses) in 472 women taken from April 2017 to June 2018 were included. Six radiologists (three inexperienced [<1 year of experience] and three experienced [10-15 years of experience]) individually reviewed US images with and without the aid of AI-CAD, first sequentially and then simultaneously. Diagnostic performance and interobserver agreement were calculated and compared between radiologists and AI-CAD. RESULTS: After implementing AI-CAD, the specificity, positive predictive value (PPV), and accuracy significantly improved, regardless of experience and workflow (all P<0.001, respectively). The overall area under the receiver operating characteristic curve significantly increased in simultaneous reading, but only for inexperienced radiologists. The agreement for Breast Imaging Reporting and Database System (BI-RADS) descriptors generally increased when AI-CAD was used (κ=0.29-0.63 to 0.35-0.73). Inexperienced radiologists tended to concede to AI-CAD results more easily than experienced radiologists, especially in simultaneous reading (P<0.001). The conversion rates for final assessment changes from BI-RADS 2 or 3 to BI-RADS higher than 4a or vice versa were also significantly higher in simultaneous reading than sequential reading (overall, 15.8% and 6.2%, respectively; P<0.001) for both inexperienced and experienced radiologists. CONCLUSION: Using AI-CAD to interpret breast US improved the specificity, PPV, and accuracy of radiologists regardless of experience level. AI-CAD may work better in simultaneous reading to improve diagnostic performance and agreement between radiologists, especially for inexperienced radiologists.

AI-CAD for differentiating lesions presenting as calcifications only on mammography: outcome analysis incorporating the ACR BI-RADS descriptors for calcifications.

OBJECTIVES: To evaluate how AI-CAD triages calcifications and to compare its performance to an experienced breast radiologist. METHODS: Among routine mammography performed between June 2016 and May 2018, 535 lesions detected as calcifications only on mammography in 500 women (mean age, 48.8 years) that were additionally interpreted with additional magnification views were included in this study. One dedicated breast radiologist retrospectively reviewed the magnification mammograms to assess morphology, distribution, and final assessment category according to ACR BI-RADS. AI-CAD analyzed routine mammograms providing AI-CAD marks and corresponding AI-CAD scores (ranging from 0 to 100%), for which values ≥ 10% were considered positive. Ground truth in terms of malignancy or benignity was confirmed with a histopathologic diagnosis or at least 1 year of imaging follow - up. RESULTS: Of the 535 calcifications, 215 (40.2%) were malignant. Calcifications with positive AI-CAD scores showed significantly higher PPVs compared to calcifications with negative scores for all morphology (all p < 0.05). PPVs were significantly higher in calcifications with positive AI-CAD scores compared to those with negative scores for BI-RADS 3, 4a, or 4b assessments (all p < 0.05). AI-CAD and the experienced radiologist did not show significant difference in diagnostic performance; sensitivity 92.1% vs 95.4% (p = 0.125), specificity 71.9% vs 72.5% (p = 0.842), and accuracy 80.0% vs 81.7% (p = 0.413). CONCLUSION: Among calcifications with same morphology or BI-RADS assessment, those with positive AI-CAD scores had significantly higher PPVs. AI-CAD showed similar diagnostic performances to the experienced radiologist for calcifications detected on mammography. KEY POINTS: • Among calcifications with same morphology or BI-RADS assessment, those with positive AI-CAD scores had significantly higher PPVs. • AI-CAD showed similar diagnostic performance to an experienced radiologist in assessing lesions detected as calcifications only on mammography. • Among malignant calcifications, calcifications with positive AI-CAD scores showed higher rates of invasive cancers than calcifications with negative scores (all p > 0.05).

Depiction of breast cancers on digital mammograms by artificial intelligence-based computer-assisted diagnosis according to cancer characteristics.

OBJECTIVE: To evaluate how breast cancers are depicted by artificial intelligence-based computer-assisted diagnosis (AI-CAD) according to clinical, radiological, and pathological factors. MATERIALS AND METHODS: From January 2017 to December 2017, 896 patients diagnosed with 930 breast cancers were enrolled in this retrospective study. Commercial AI-CAD was applied to digital mammograms and abnormality scores were obtained. We evaluated the abnormality score according to clinical, radiological, and pathological characteristics. False-negative results were defined by abnormality scores less than 10. RESULTS: The median abnormality score of 930 breasts was 87.4 (range 0-99). The false-negative rate of AI-CAD was 19.4% (180/930). Cancers with an abnormality score of more than 90 showed a high proportion of palpable lesions, BI-RADS 4c and 5 lesions, cancers presenting as mass with or without microcalcifications and invasive cancers compared with low-scored cancers (all p < 0.001). False-negative cancers were more likely to develop in asymptomatic patients and extremely dense breasts and to be diagnosed as occult breast cancers and DCIS compared to detected cancers. CONCLUSION: Breast cancers depicted with high abnormality scores by AI-CAD are associated with higher BI-RADS category, invasive pathology, and higher cancer stage. KEY POINTS: • High-scored cancers by AI-CAD included a high proportion of BI-RADS 4c and 5 lesions, masses with or without microcalcifications, and cancers with invasive pathology. • Among invasive cancers, cancers with higher T and N stage and HER2-enriched subtype were depicted with higher abnormality scores by AI-CAD. • Cancers missed by AI-CAD tended to be in asymptomatic patients and extremely dense breasts and to be diagnosed as occult breast cancers by radiologists.