Evaluation of an artificial intelligence model for identification of mass effect and vasogenic edema on computed tomography of the head.

BACKGROUND AND PURPOSE: Mass effect and vasogenic edema are critical findings on CT of the head. This study compared the accuracy of an artificial intelligence model (Annalise Enterprise CTB) to consensus neuroradiologist interpretations in detecting mass effect and vasogenic edema. MATERIALS AND METHODS: A retrospective standalone performance assessment was conducted on datasets of non-contrast CT head cases acquired between 2016 and 2022 for each finding. The cases were obtained from patients aged 18 years or older from five hospitals in the United States. The positive cases were selected consecutively based on the original clinical reports using natural language processing and manual confirmation. The negative cases were selected by taking the next negative case acquired from the same CT scanner after positive cases. Each case was interpreted independently by up to three neuroradiologists to establish consensus interpretations. Each case was then interpreted by the AI model for the presence of the relevant finding. The neuroradiologists were provided with the entire CT study. The AI model separately received thin (≤1.5mm) and/or thick (>1.5 and ≤5mm) axial series. RESULTS: The two cohorts included 818 cases for mass effect and 310 cases for vasogenic edema. The AI model identified mass effect with sensitivity 96.6% (95% CI, 94.9-98.2) and specificity 89.8% (95% CI, 84.7-94.2) for the thin series, and 95.3% (95% CI, 93.5-96.8) and 93.1% (95% CI, 89.1-96.6) for the thick series. It identified vasogenic edema with sensitivity 90.2% (95% CI, 82.0-96.7) and specificity 93.5% (95% CI, 88.9-97.2) for the thin series, and 90.0% (95% CI, 84.0-96.0) and 95.5% (95% CI, 92.5-98.0) for the thick series. The corresponding areas under the curve were at least 0.980. CONCLUSIONS: The assessed AI model accurately identified mass effect and vasogenic edema in this CT dataset. It could assist the clinical workflow by prioritizing interpretation of abnormal cases, which could benefit patients through earlier identification and subsequent treatment. ABBREVIATIONS: AI = artificial intelligence; AUC = area under the curve; CADt = computer assisted triage devices; FDA = Food and Drug Administration; NPV = negative predictive value; PPV = positive predictive value; SD = standard deviation.

Missed Screening Mammography Appointments: Patient Sociodemographic Characteristics and Mammography Completion After 1 Year.

OBJECTIVE: Patients who miss screening mammogram appointments without notifying the health care system (no-show) risk care delays. We investigate sociodemographic characteristics of patients who experience screening mammogram no-shows at a community health center and whether and when the missed examinations are completed. METHODS: We included patients with screening mammogram appointments at a community health center between January 1, 2021, and December 31, 2021. Language, race, ethnicity, insurance type, residential ZIP code tabulation area (ZCTA) poverty, appointment outcome (no-show, same-day cancelation, completed), and dates of completed screening mammograms after no-show appointments with ≥1-year follow-up were collected. Multivariable analyses were used to assess associations between patient characteristics and appointment outcomes. RESULTS: Of 6,159 patients, 12.1% (743 of 6,159) experienced no-shows. The no-show group differed from the completed group by language, race and ethnicity, insurance type, and poverty level (all P < .05). Patients with no-shows more often had: primary language other than English (32.0% [238 of 743] versus 26.7% [1,265 of 4,741]), race and ethnicity other than White non-Hispanic (42.3% [314 of 743] versus 33.6% [1,595 of 4,742]), Medicaid or means-tested insurance (62.0% [461 of 743] versus 34.4% [1,629 of 4,742]), and residential ZCTAs with ≥20% poverty (19.5% [145 of 743] versus 14.1% [670 of 4,742]). Independent predictors of no-shows were Black non-Hispanic race and ethnicity (adjusted odds ratio [aOR], 1.52; 95% confidence interval [CI], 1.12-2.07; P = .007), Medicaid or other means-tested insurance (aOR, 2.75; 95% CI, 2.29-3.30; P < .001), and ZCTAs with ≥20% poverty (aOR, 1.76; 95% CI, 1.14-2.72; P = .011). At 1-year follow-up, 40.6% (302 of 743) of patients with no-shows had not completed screening mammogram. DISCUSSION: Screening mammogram no-shows is a health equity issue in which socio-economically disadvantaged and racially and ethnically minoritized patients are more likely to experience missed appointments and continued delays in screening mammogram completion.

Incidence, Timing, and Long-Term Outcomes of COVID-19 Vaccine-Related Lymphadenopathy on Screening Mammography.

OBJECTIVE: To determine the incidence, timing, and long-term outcomes of unilateral axillary lymphadenopathy ipsilateral to vaccine site (UIAL) on screening mammography after COVID-19 vaccination. METHODS: This retrospective, multisite study included consecutive patients undergoing screening mammography February 8, 2021, to January 31, 2022, with at least 1 year of follow-up. UIAL was typically considered benign (BI-RADS 1 or 2) in the setting of recent (≤6 weeks) vaccination or BI-RADS 0 (ultrasound recommended) when accompanied by a breast finding or identified >6 weeks postvaccination. Vaccination status and manufacturer were obtained from regional registries. Lymphadenopathy rates in vaccinated patients with and without UIAL were compared using Pearson's χ2 test. RESULTS: There were 44,473 female patients (mean age 60.4 ± 11.4 years) who underwent screening mammography at five sites, and 40,029 (90.0%) received at least one vaccine dose. Ninety-four (0.2%) presented with UIAL, 1 to 191 days postvaccination (median 13.5 [interquartile range: 5.0-31.0]). Incidence declined from 2.1% to 0.9% to ≤0.5% after 1, 2, and 3 weeks and persisted up to 36 weeks (P < .001). UIAL did not vary across manufacturer (P = .15). Of 94, 77 (81.9%) were BI-RADS 1 or 2 at screening. None were diagnosed with malignancy at 1-year follow-up. Seventeen (18.1%) were BI-RADS 0 at screening. At diagnostic workup, 13 (76.5%) were BI-RADS 1 or 2, 2 (11.8%) were BI-RADS 3, and 2 (11.8%) were BI-RADS 4. Both BI-RADS 4 patients had malignant status and ipsilateral breast malignancies. Of BI-RADS 3 patients, at follow-up, one was biopsied yielding benign etiology, and one was downgraded to BI-RADS 2. DISCUSSION: Isolated UIAL on screening mammography performed within 6 months of COVID-19 vaccination can be safely assessed as benign.

Society of Radiologists in Ultrasound Consensus Conference Recommendations for Incidental Gallbladder Polyp Management: Interreader Agreement Among 10 Radiologists.

BACKGROUND. The 2022 Society of Radiologists in Ultrasound (SRU) consensus conference recommendations for small gallbladder polyps support management that is less aggressive than earlier approaches and may help standardize evaluation of polyps by radiologists. OBJECTIVE. The purpose of the present study was to assess the interreader agreement of radiologists in applying SRU recommendations for management of incidental gallbladder polyps on ultrasound. METHODS. This retrospective study included 105 patients (75 women and 30 men; median age, 51 years) with a gallbladder polyp on ultrasound (without features highly suspicious for invasive or malignant tumor) who underwent cholecystectomy between January 1, 2003, and January 1, 2021. Ten abdominal radiologists independently reviewed ultrasound examinations and, using the SRU recommendations, assessed one polyp per patient to assign risk category (extremely low risk, low risk, or indeterminate risk) and make a possible recommendation for surgical consultation. Five radiologists were considered less experienced (< 5 years of experience), and five were considered more experienced (≥ 5 years of experience). Interreader agreement was evaluated. Polyps were classified pathologically as nonneoplastic or neoplastic. RESULTS. For risk category assignments, interreader agreement was substantial among all readers (k = 0.710), less-experienced readers (k = 0.705), and more-experienced readers (k = 0.692). For surgical consultation recommendations, inter-reader agreement was substantial among all readers (k = 0.795) and more-experienced readers (k = 0.740) and was almost perfect among less-experienced readers (k = 0.811). Of 10 readers, a median of 5.0 (IQR, 2.0-8.0), 4.0 (IQR, 2.0-7.0), and 0.0 (IQR, 0.0-0.0) readers classified polyps as extremely low risk, low risk, and indeterminate risk, respectively. Across readers, the percentage of polyps classified as extremely low risk ranged from 32% to 72%; as low risk, from 24% to 65%; and as indeterminate risk, from 0% to 8%. Of 10 readers, a median of zero change to 0 (IQR, 0.0-1.0) readers recommended surgical consultation; the percentage of polyps receiving a recommendation for surgical consultation ranged from 4% to 22%. Of a total of 105 polyps, 102 were nonneo-plastic and three were neoplastic (all benign). Based on readers' most common assessments for nonneoplastic polyps, the risk category was extremely low risk for 53 polyps, low risk for 48 polyps, and indeterminate risk for one polyp; surgical consultation was recommended for 16 polyps. CONCLUSION. Ten abdominal radiologists showed substantial agreement for polyp risk categorizations and surgical consultation recommendations, although areas of reader variability were identified. CLINICAL IMPACT. The findings support the overall reproducibility of the SRU recommendations, while indicating opportunity for improvement.

Performance Metrics of Screening Digital Breast Tomosynthesis Based on Years Since a Prior Breast Cancer Diagnosis.

BACKGROUND. Mammography surveillance protocols after breast cancer treatment vary widely. Some practices recommend performing diagnostic mammography for a certain number of years or indefinitely, whereas others recommend returning immediately to screening. OBJECTIVE. This study's objective was to determine performance metrics of screening digital breast tomosynthesis (DBT) in patients who resume screening mammography immediately after breast cancer treatment, based on the number of years since the breast cancer diagnosis. METHODS. This retrospective study included screening DBT examinations performed from January 2013 to June 2019 in patients who resumed screening mammography immediately after a prior breast cancer diagnosis. Multivariable logistic regression models with generalized estimating equations were used to evaluate associations between screening performance metrics and years since the prior breast cancer diagnosis, controlling for age, race and ethnicity, breast density, presence of a prior screening mammogram, and interpreting radiologist. RESULTS. The study included 8090 patients (mean age, 65 ± 11 [SD] years) with a prior breast cancer diagnosis who underwent 30,812 screening DBT examinations during the study period. The cancer detection rate (CDR) was 8.6 per 1000 examinations (265/30,812), abnormal interpretation rate (AIR) was 5.7% (1750/30,812), PPV1 was 15.1% (265/1750), sensitivity was 80.3% (265/330), specificity was 95.1% (28,997/30,482), and false-negative rate was 2.1 per 1000 examinations (65/30,812). CDR showed a significant independent positive association with years since breast cancer diagnosis (adjusted OR, 1.03; 95% CI, 1.01-1.05; p < .001), being lowest more than 2 to up to 3 years after diagnosis (4.9 per 1000 examinations) and highest more than 8 to up to 9 years after diagnosis (11.2 per 1000 examinations). AIR showed a significant independent negative association with years since breast cancer diagnosis (adjusted OR, 0.99; 95% CI, 0.98-1.00; p = .01), being highest 1 year or less after diagnosis (7.5%) and lowest more than 5 to up to 6 years after diagnosis (5.0%). CONCLUSION. Among 8090 patients with a prior breast cancer diagnosis, even though the AIR was higher during the year after diagnosis compared with subsequent years, the AIR remained acceptably low (< 10%) in all years. CLINICAL IMPACT. These results support the study institution's mammographic surveillance protocol for patients with a prior breast cancer diagnosis of returning immediately to DBT screening.

Prediction of Surgical Upstaging Risk of Ductal Carcinoma In Situ Using Machine Learning Models.

Objective: The purpose of this study was to build machine learning models to predict surgical upstaging risk of ductal carcinoma in situ (DCIS) to invasive cancer and to compare model performance to eligibility criteria used by the Comparison of Operative versus Monitoring and Endocrine Therapy (COMET) active surveillance trial. Methods: Medical records were retrospectively reviewed of all women with DCIS at core-needle biopsy who underwent surgery from 2007 to 2016 at an academic medical center. Multivariable regression and machine learning models were developed to evaluate upstaging-related features and their performance was compared with that achieved using the COMET trial eligibility criteria. Results: Of 1387 women (mean age, 57 years; range, 27-89 years), the upstaging rate of DCIS was 17% (235/1387). On multivariable analysis, upstaging-associated features were presentation of DCIS as a palpable area of concern, imaging finding of a mass, and nuclear grades 2 or 3 at biopsy (P < 0.05). If COMET trial eligibility criteria were applied to our study cohort, then 496 women (42%, 496/1175) would have been eligible for the trial, with an upstaging rate of 12% (61/496). Of the machine learning models, none had a significantly lower upstaging rate than 12%. However, if using the models to determine eligibility, then a significantly larger proportion of women (56%-87%) would have been eligible for active surveillance. Conclusion: Use of machine learning models to determine eligibility for the COMET trial identified a larger proportion of women eligible for surveillance compared with current eligibility criteria while maintaining similar upstaging rates.

Comparison of the Diagnostic Accuracy of Mammogram-based Deep Learning and Traditional Breast Cancer Risk Models in Patients Who Underwent Supplemental Screening with MRI.

Background Access to supplemental screening breast MRI is determined using traditional risk models, which are limited by modest predictive accuracy. Purpose To compare the diagnostic accuracy of a mammogram-based deep learning (DL) risk assessment model to that of traditional breast cancer risk models in patients who underwent supplemental screening with MRI. Materials and Methods This retrospective study included consecutive patients undergoing breast cancer screening MRI from September 2017 to September 2020 at four facilities. Risk was assessed using the Tyrer-Cuzick (TC) and National Cancer Institute Breast Cancer Risk Assessment Tool (BCRAT) 5-year and lifetime models as well as a DL 5-year model that generated a risk score based on the most recent screening mammogram. A risk score of 1.67% or higher defined increased risk for traditional 5-year models, a risk score of 20% or higher defined high risk for traditional lifetime models, and absolute scores of 2.3 or higher and 6.6 or higher defined increased and high risk, respectively, for the DL model. Model accuracy metrics including cancer detection rate (CDR) and positive predictive values (PPVs) (PPV of abnormal findings at screening [PPV1], PPV of biopsies recommended [PPV2], and PPV of biopsies performed [PPV3]) were compared using logistic regression models. Results This study included 2168 women who underwent 4247 high-risk screening MRI examinations (median age, 54 years [IQR, 48-60 years]). CDR (per 1000 examinations) was higher in patients at high risk according to the DL model (20.6 [95% CI: 11.8, 35.6]) than according to the TC (6.0 [95% CI: 2.9, 12.3]; P < .01) and BCRAT (6.8 [95% CI: 2.9, 15.8]; P = .04) lifetime models. PPV1, PPV2, and PPV3 were higher in patients identified as high risk by the DL model (PPV1, 14.6%; PPV2, 32.4%; PPV3, 36.4%) than those identified as high risk with the TC (PPV1, 5.0%; PPV2, 12.7%; PPV3, 13.5%; P value range, .02-.03) and BCRAT (PPV1, 5.5%; PPV2, 11.1%; PPV3, 12.5%; P value range, .02-.05) lifetime models. Conclusion Patients identified as high risk by a mammogram-based DL risk assessment model showed higher CDR at breast screening MRI than patients identified as high risk with traditional risk models. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Bae in this issue.

Impact of Baseline Interstitial Lung Abnormalities on Pneumonitis Risk in Patients Receiving Immune Checkpoint Inhibitors for Non-Small-Cell Lung Cancer.

INTRODUCTION/BACKGROUND: Immune-related pneumonitis is a potentially fatal complication of treatment with immune checkpoint inhibitors (ICIs). Interstitial lung disease (ILD) is associated with increased risk for pneumonitis, but the impact of interstitial abnormalities (ILA) in the absence of ILD has not been extensively assessed. We examined the relationship between ILA on pretreatment chest computed tomography (CT) scans and risk of pneumonitis in patients with non-small-cell lung cancer (NSCLC). METHODS: This retrospective cohort study included consecutive adult patients who received ICI for NSCLC between January 2013 and January 2020 at our institution. Two thoracic radiologists blinded to clinical outcomes independently reviewed pre-ICI chest CTs to identify and categorize ILA using previously published definitions. We used uni- and multivariable analysis adjusted for age, radiation, and smoking status to assess for associations between ILA, clinicopathologic characteristics, and symptomatic (CTCAE grade ≥2) pneumonitis. RESULTS: Of 475 patients who received ICI treatment and met inclusion criteria, baseline ILA were present in 78 (16.4%) patients, most commonly as a subpleural nonfibrotic pattern. In total, 43 (9.1%) of 475 patients developed symptomatic pneumonitis. Pneumonitis occurred in 16.7% of patients with ILA compared to 7.6% patients without ILA (P < .05). Presence of ground glass and extent of lung parenchymal involvement were associated with an increased risk of pneumonitis. On multivariable analysis, baseline ILA remained associated with increased risk of symptomatic pneumonitis (OR 2.2, 95% CI, 1.0-4.5). CONCLUSIONS: Baseline ILAs are associated with the development of symptomatic pneumonitis in patients with NSCLC treated with ICI. Additional studies are needed to validate these observations.

Cardiac Incidental Findings on Abdominopelvic Computed Tomography: Prevalence and Association with Subsequent Cardiovascular Events.

PURPOSE: The aim of this study was to assess the prevalence of reportable cardiac findings detected on abdominopelvic CTs and the association with subsequent cardiovascular events. MATERIALS AND METHODS: We performed a retrospective search of electronic medical record of patients who underwent abdominopelvic CT between November 2006 and November 2011 with a clinical history of upper abdominal pain. A radiologist blinded to the original CT report reviewed all 222 cases for the presence of pertinent reportable cardiac findings. The original CT report was also evaluated for documentation of pertinent reportable cardiac findings. The following findings were recorded on all CTs: presence of coronary calcification, fatty metaplasia, ventricle wall thinning and thickening, valve calcification or prosthesis, heart/chamber enlargement, aneurysm, mass, thrombus, device, air within ventricles, abnormal pericardium, prior sternotomy, and adhesions if prior sternotomy. Medical records were reviewed to identify cardiovascular events on follow-up in patients with the presence or absence of cardiac findings. We compared the distribution findings in patients with and without cardiac events using the Wilcoxon test (for continuous variables) and the Pearson's chi-squared test (for categorical variables). RESULTS: Eighty-five of 222 (38.3%) patients (52.7% females, median age 52.5 years) had at least one pertinent reportable cardiac finding on the abdominopelvic CT, with a total of 140 findings in this group. From the total 140 findings, 100 (71.4%) were not reported. The most common findings seen on abdominal CTs were: coronary artery calcification (66 patients), heart or chamber enlargement (25), valve abnormality (19), sternotomy and surgery signs (9), LV wall thickening (7), device (5), LV wall thinning (2), pericardial effusion (5), and others (3). After a mean follow-up of 43.9 months, 19 cardiovascular events were found in the cohort (transient ischemic attack, cerebrovascular accident, myocardial infarction, cardiac arrest, acute arrhythmia, palpitation, syncope and acute chest pain). Only 1 event occurred in the group of patients with no incidental pertinent reportable cardiac findings (1/137 = 0.73%). All other 18 events occurred in patients with incidental pertinent reportable cardiac findings (18/85 = 21.2%), which was significantly different (p < 0.0001). One out of the total 19 events in the overall group (5.24%) occurred in a patient with no incidental pertinent reportable cardiac findings while 18 of 19 total events (94.74%) occurred with patients with incidental pertinent reportable cardiac findings, which was also significantly different (p < 0.001). Fifteen of the total events (79%) occurred in patients in whom the incidental pertinent reportable cardiac findings were not reported, which was significantly different (p < 0.001) from the four events that occurred in patients in whom the incidental pertinent reportable cardiac findings were reported or had no findings. CONCLUSIONS: Incidental pertinent reportable cardiac findings are common on abdominal CTs and are frequently not reported by radiologists. These findings are of clinical relevance since patients with pertinent reportable cardiac findings have a significantly higher incidence of cardiovascular events on follow-up.

Assessing the Relationship Between Race, Language, and Surgical Admissions in the Emergency Department.

INTRODUCTION: English proficiency and race are both independently known to affect surgical access and quality, but relatively little is known about the impact of race and limited English proficiency (LEP) on admission for emergency surgery from the emergency department (ED). Our objective was to examine the influence of race and English proficiency on admission for emergency surgery from the ED. METHODS: We conducted a retrospective observational cohort study from January 1-December 31, 2019 at a large, quaternary-care urban, academic medical center with a 66-bed ED Level I trauma and burn center. We included ED patients of all self-reported races reporting a preferred language other than English and requiring an interpreter or declaring English as their preferred language (control group). A multivariable logistic regression was fit to assess the association of LEP status, race, age, gender, method of arrival to the ED, insurance status, and the interaction between LEP status and race with admission for surgery from the ED. RESULTS: A total of 85,899 patients (48.1% female) were included in this analysis, of whom 3,179 (3.7%) were admitted for emergent surgery. Regardless of LEP status, patients identifying as Black (odds ratio [OR] 0.456, 95% CI 0.388-0.533; P<0.005), Asian [OR 0.759, 95% CI 0.612-0.929]; P=0.009), or female [OR 0.926, 95% CI 0.862-0.996]; P=0.04) had significantly lower odds for admission for surgery from the ED compared to White patients. Compared to individuals on Medicare, those with private insurance [OR 1.25, 95% CI 1.13-1.39; P <0.005) were significantly more likely to be admitted for emergent surgery, whereas those without insurance [OR 0.581, 95% CI 0.323-0.958; P=0.05) were significantly less likely to be admitted for emergent surgery. There was no significant difference in odds of admission for surgery between LEP vs non-LEP patients. CONCLUSION: Individuals without health insurance and those identifying as female, Black, or Asian had significantly lower odds of admission for surgery from the ED compared to those with health insurance, males, and those self-identifying as White, respectively. Future studies should assess the reasons underpinning this finding to elucidate impact on patient outcomes.

Improving Prostate Cancer Detection With MRI: A Multi-Reader, Multi-Case Study Using Computer-Aided Detection (CAD).

RATIONALE AND OBJECTIVES: To evaluate whether addition of a computer-aided diagnostic (CAD) generated MRI series improves detection of clinically significant prostate cancer. MATERIALS AND METHODS: Nine radiologists retrospectively interpreted 150 prostate MRI examinations without and then with an additional random forest-based CAD model-generated MRI series. Characteristics of biopsy negative versus positive (Gleason ≥ 7 adenocarcinoma) groups were compared using the Wilcoxon test for continuous and Pearson's chi-squared test for categorical variables. The diagnostic performance of readers was compared without versus with CAD using MRMC methods to estimate the area under the receiver operator characteristic curve (AUC). Inter-reader agreement was assessed using weighted inter-rater agreement statistics. Analyses were repeated in peripheral and transition zone subgroups. RESULTS: Among 150 men with median age 67 ± 7.4 years, those with clinically significant prostate cancer were older (68 ± 7.6 years vs. 66 ± 7.0 years; p < .02), had smaller prostate volume (43.9 mL vs. 60.6 mL; p < .001), and no difference in prostate specific antigen (PSA) levels (7.8 ng/mL vs. 6.9 ng/mL; p = .08), but higher PSA density (0.17 ng/mL/cc vs. 0.10 ng/mL/cc; p < .001). Inter-rater agreement (IRA) for PI-RADS scores was moderate without CAD and significantly improved to substantial with CAD (IRA = 0.47 vs. 0.65; p < .001). CAD also significantly improved average reader AUC (AUC = 0.72, vs. AUC = 0.67; p = .02). CONCLUSION: Addition of a random forest method-based, CAD-generated MRI image series improved inter-reader agreement and diagnostic performance for detection of clinically significant prostate cancer, particularly in the transition zone.

Evaluation of an Artificial Intelligence Model for Detection of Pneumothorax and Tension Pneumothorax in Chest Radiographs.

Importance: Early detection of pneumothorax, most often via chest radiography, can help determine need for emergent clinical intervention. The ability to accurately detect and rapidly triage pneumothorax with an artificial intelligence (AI) model could assist with earlier identification and improve care. Objective: To compare the accuracy of an AI model vs consensus thoracic radiologist interpretations in detecting any pneumothorax (incorporating both nontension and tension pneumothorax) and tension pneumothorax. Design, Setting, and Participants: This diagnostic study was a retrospective standalone performance assessment using a data set of 1000 chest radiographs captured between June 1, 2015, and May 31, 2021. The radiographs were obtained from patients aged at least 18 years at 4 hospitals in the Mass General Brigham hospital network in the United States. Included radiographs were selected using 2 strategies from all chest radiography performed at the hospitals, including inpatient and outpatient. The first strategy identified consecutive radiographs with pneumothorax through a manual review of radiology reports, and the second strategy identified consecutive radiographs with tension pneumothorax using natural language processing. For both strategies, negative radiographs were selected by taking the next negative radiograph acquired from the same radiography machine as each positive radiograph. The final data set was an amalgamation of these processes. Each radiograph was interpreted independently by up to 3 radiologists to establish consensus ground-truth interpretations. Each radiograph was then interpreted by the AI model for the presence of pneumothorax and tension pneumothorax. This study was conducted between July and October 2021, with the primary analysis performed between October and November 2021. Main Outcomes and Measures: The primary end points were the areas under the receiver operating characteristic curves (AUCs) for the detection of pneumothorax and tension pneumothorax. The secondary end points were the sensitivities and specificities for the detection of pneumothorax and tension pneumothorax. Results: The final analysis included radiographs from 985 patients (mean [SD] age, 60.8 [19.0] years; 436 [44.3%] female patients), including 307 patients with nontension pneumothorax, 128 patients with tension pneumothorax, and 550 patients without pneumothorax. The AI model detected any pneumothorax with an AUC of 0.979 (95% CI, 0.970-0.987), sensitivity of 94.3% (95% CI, 92.0%-96.3%), and specificity of 92.0% (95% CI, 89.6%-94.2%) and tension pneumothorax with an AUC of 0.987 (95% CI, 0.980-0.992), sensitivity of 94.5% (95% CI, 90.6%-97.7%), and specificity of 95.3% (95% CI, 93.9%-96.6%). Conclusions and Relevance: These findings suggest that the assessed AI model accurately detected pneumothorax and tension pneumothorax in this chest radiograph data set. The model's use in the clinical workflow could lead to earlier identification and improved care for patients with pneumothorax.

Predicting ipsilateral recurrence in women treated for ductal carcinoma in situ using machine learning and multivariable logistic regression models.

PURPOSE: To develop machine learning (ML) and multivariable regression models to predict ipsilateral breast event (IBE) risk after ductal carcinoma in situ (DCIS) treatment. METHODS: A retrospective investigation was conducted of patients diagnosed with DCIS from 2007 to 2014 who were followed for a minimum of five years after treatment. Data about each patient were extracted from the medical records. Two ML models (penalized logistic regression and random forest) and a multivariable logistic regression model were developed to evaluate recurrence-related variables. RESULTS: 650 women (mean age 56 years, range 27-87 years) underwent treatment for DCIS and were followed for at least five years after treatment (mean 8.0 years). 5.5% (n = 36) experienced an IBE. With multivariable analysis, the variables associated with higher IBE risk were younger age (adjusted odds ratio [aOR] 0.96, p = 0.02), dense breasts at mammography (aOR 3.02, p = 0.02), and < 5 years of endocrine therapy (aOR 4.48, p = 0.02). The multivariable regression model to predict IBE risk achieved an area under the receiver operating characteristic curve (AUC) of 0.75 (95% CI 0.67-0.84). The penalized logistic regression and random forest models achieved mean AUCs of 0.52 (95% CI 0.42-0.61) and 0.54 (95% CI 0.43-0.65), respectively. CONCLUSION: Variables associated with higher IBE risk after DCIS treatment include younger age, dense breasts, and <5 years of adjuvant endocrine therapy. The multivariable logistic regression model attained the highest AUC (0.75), suggesting that regression models have a critical role in risk prediction for patients with DCIS.

Deep Learning vs Traditional Breast Cancer Risk Models to Support Risk-Based Mammography Screening.

BACKGROUND: Deep learning breast cancer risk models demonstrate improved accuracy compared with traditional risk models but have not been prospectively tested. We compared the accuracy of a deep learning risk score derived from the patient's prior mammogram to traditional risk scores to prospectively identify patients with cancer in a cohort due for screening. METHODS: We collected data on 119 139 bilateral screening mammograms in 57 617 consecutive patients screened at 5 facilities between September 18, 2017, and February 1, 2021. Patient demographics were retrieved from electronic medical records, cancer outcomes determined through regional tumor registry linkage, and comparisons made across risk models using Wilcoxon and Pearson χ2 2-sided tests. Deep learning, Tyrer-Cuzick, and National Cancer Institute Breast Cancer Risk Assessment Tool (NCI BCRAT) risk models were compared with respect to performance metrics and area under the receiver operating characteristic curves. RESULTS: Cancers detected per thousand patients screened were higher in patients at increased risk by the deep learning model (8.6, 95% confidence interval [CI] = 7.9 to 9.4) compared with Tyrer-Cuzick (4.4, 95% CI = 3.9 to 4.9) and NCI BCRAT (3.8, 95% CI = 3.3 to 4.3) models (P < .001). Area under the receiver operating characteristic curves of the deep learning model (0.68, 95% CI = 0.66 to 0.70) was higher compared with Tyrer-Cuzick (0.57, 95% CI = 0.54 to 0.60) and NCI BCRAT (0.57, 95% CI = 0.54 to 0.60) models. Simulated screening of the top 50th percentile risk by the deep learning model captured statistically significantly more patients with cancer compared with Tyrer-Cuzick and NCI BCRAT models (P < .001). CONCLUSIONS: A deep learning model to assess breast cancer risk can support feasible and effective risk-based screening and is superior to traditional models to identify patients destined to develop cancer in large screening cohorts.

Long-term Cardiovascular Risk in Astronauts: Comparing NASA Mission Astronauts With a Healthy Cohort From the Cooper Center Longitudinal Study.

OBJECTIVE: To determine the long-term cardiovascular disease risk of astronauts with spaceflight exposure compared with a well-matched cohort. METHODS: National Aeronautics and Space Administration (NASA) astronauts are selected into their profession based upon education, unique skills, and health and are exposed to cardiovascular disease risk factors during spaceflight. The Cooper Center Longitudinal Study (CCLS) is a generally healthy cohort from a preventive medicine clinic in Dallas, Texas. Using a matched cohort design, astronauts who were selected beginning April 1, 1959, (and each subsequent selection class through 2009) and exposed to spaceflight were matched to CCLS participants who met astronaut selection criteria; 1514 CCLS participants matched to 303 astronauts in a 5-to-1 ratio on sex, date of birth, and age. The outcome of cardiovascular mortality through December 31, 2016, was determined by death certificate or National Death Index. RESULTS: There were 11 deaths caused by cardiovascular disease (CVD) among astronauts and 46 among CCLS participants. There was no evidence of increased mortality risk in astronauts (hazard ratio [HR]=1.10; 95% confidence interval [CI], 0.50 to 2.45) with adjustment for baseline cardiovascular covariates. However, the secondary outcome of CVD events showed an increased adjusted risk in astronauts (HR=2.41; 95% CI, 1.26 to 4.63). CONCLUSION: No increased risk of CVD mortality was observed in astronauts with spaceflight exposure compared with a well-matched cohort, but there was evidence of increased total CVD events. Given that the duration of spaceflight will increase, particularly on missions to Mars, continued surveillance and mitigation of CVD risk is needed to ensure the safety of those who venture into space.

Surveillance for jugular venous thrombosis in astronauts.

BACKGROUND: Thrombosis of the left internal jugular vein in an astronaut aboard the International Space Station was recently described, incidentally discovered during a research study of blood flow in neck veins in microgravity. Given this event, and the high incidence of flow abnormalities, the National Aeronautics and Space Administration (NASA) instituted an occupational surveillance program to evaluate astronauts for venous thrombosis. METHODS: Duplex ultrasound of the bilateral internal jugular veins was conducted on all NASA astronauts terrestrially, and at three points during spaceflight. Respiratory maneuvers were performed. Images were analyzed for thrombosis and certain hemodynamic characteristics, including peak velocity and degree of echogenicity. RESULTS: Eleven astronauts were evaluated with matching terrestrial and in-flight ultrasounds. No thrombosis was detected. Compared to terrestrial ultrasound measurements, in-flight peak velocity was reduced and lowest in the left. Six of 11 astronauts had mild-moderate echogenicity in the left internal jugular vein during spaceflight, but none had more than mild echogenicity in the right internal jugular vein. Two astronauts developed retrograde blood flow in the left internal jugular vein. CONCLUSION: Abnormal flow characteristics in microgravity, most prominent in the left internal jugular vein, may signal an increased risk for thrombus formation in some individuals.

Impact of a Deep Learning Model for Predicting Mammographic Breast Density in Routine Clinical Practice.

OBJECTIVE: Legislation in 38 states requires patient notification of dense mammographic breast tissue because increased density is a marker of breast cancer risk and can limit mammographic sensitivity. Because radiologist density assessments vary widely, our objective was to implement and measure the impact of a deep learning (DL) model on mammographic breast density assessments in clinical practice. METHODS: This institutional review board-approved prospective study identified consecutive screening mammograms performed across three clinical sites over two periods: 2017 period (January 1, 2017, through September 30, 2017) and 2019 period (January 1, 2019, through September 30, 2019). The DL model was implemented at sites A (academic practice) and B (community practice) in 2018 for all screening mammograms. Site C (community practice) was never exposed to the DL model. Prospective densities were evaluated, and multivariable logistic regression models evaluated the odds of a dense mammogram classification as a function of time and site. RESULTS: We identified 85,124 consecutive screening mammograms across the three sites. Across time intervals, odds of a dense classification decreased at sites exposed to the DL model, site A (adjusted odds ratio [aOR], 0.93; 95% confidence interval [CI], 0.86-0.99; P = .024) and site B (aOR, 0.81 [95% CI, 0.70-0.93]; P = .003), and odds increased at the site unexposed to the model (site C) (aOR, 1.13 [95% CI, 1.01-1.27]; P = .033). DISCUSSION: A DL model reduces the odds of screening mammograms categorized as dense. Accurate density assessments could help health care systems more appropriately use limited supplemental screening resources and help better inform traditional clinical risk models.

Impact of a Same-Day Breast Biopsy Program on Disparities in Time to Biopsy for Patients With Serious Mental Illness.

PURPOSE: The aim of this study was to investigate disparities in time between breast biopsy recommendation and completion and the impact of a same-day biopsy (SDB) program for patients with serious mental illness (SMI), with a focus on more vulnerable individuals with public payer insurance. METHODS: In August 2017, the authors' academic breast imaging center started routinely offering needle biopsies on the day of recommendation. Primary outcomes were the proportion of biopsies performed as SDBs and days from biopsy recommendation to completion over a 2.5-year pre- versus postintervention period, comparing all patients with SMI versus those without, and public payer-insured patients <65 years of age with SMI (SMI-PP) versus without SMI (non-SMI-PP). Multivariable proportional odds and logistic regression models were fit to assess association of SMI status, age, race/ethnicity, language, and insurance with days to biopsy and SDB within each period. RESULTS: There were 2,026 biopsies preintervention and 2,361 biopsies postintervention. Preintervention, 8.43% of patients with SMI (7 of 83) underwent SDB compared with 15.59% of those without SMI (303 of 1,943) (P = .076), and 2.7% of the SMI-PP subgroup (1 of 37) underwent SDB compared with 15.88% of the non-SMI-PP subgroup (47 of 296) (P = .031). Adjusted for age, race/ethnicity, and language, disparities persisted in odds for undergoing SDB (adjusted odds ratio, 0.13; 95% confidence interval, 0.02-0.92; P = .04) and having longer days to biopsy (adjusted odds ratio, 2.35; 95% confidence interval, 1.26-4.37; P = .01) for the SMI-PP subgroup compared with the non-SMI-PP subgroup in the preintervention period. There was no evidence of these disparities postintervention for the SMI-PP subgroup. SDB proportion increased from 15.3% (310 of 2,026) to 36.09% (852 of 2,361) (P < .001) across all patients. CONCLUSIONS: A same-day breast biopsy program mitigates disparities in time to biopsy for patients with SMI and helps improve breast cancer care equity for this vulnerable population.

Contralateral breast cancer after curative-intent treatment for ductal carcinoma in situ: Rate and associated clinicopathological and imaging risk factors.

PURPOSE: Patients who have ductal carcinoma in situ (DCIS) are undergoing bilateral mastectomy at increasing rates. One of the reasons is to minimize contralateral breast cancer (CBC) risk. The purpose of this study is to determine the rate of and risk factors associated with CBC in women treated for DCIS. METHODS: A retrospective study was performed of women with DCIS at surgery from 2007 to 2014 who had at least five-year follow-up. Patient attributes, imaging findings, histopathology results, and surgical and long-term outcomes were collected. Features associated with a CBC were assessed with multivariable logistic regression models. RESULTS: 613 women (mean 56 years, range 30-87) with DCIS underwent breast-conserving surgery (BCS) (n = 426), unilateral mastectomy (n = 101), or bilateral mastectomy (n = 86), with mean follow-up of 7.9 years. Of the 527 women who had BCS or unilateral mastectomy, 7.4% (n = 39) developed a CBC (DCIS in 12 and invasive cancer in 27). 4.1% (5/122) of women treated with adjuvant endocrine therapy developed a CBC, compared to 8.4% (34/405) who were not treated (p = .11). Features associated with CBC risk were younger age at menarche (adjusted odds ratio [aOR] of 0.76, p = .03) and low nuclear grade of DCIS (aOR of 5.43 for grade 1 versus 3, p = .01). CONCLUSION: In women treated for DCIS, the overall rate of CBC was low at 7.4%. Younger age at menarche and low nuclear grade of DCIS had significant associations with higher CBC risk.

Multilevel Body Composition Analysis on Chest Computed Tomography Predicts Hospital Length of Stay and Complications After Lobectomy for Lung Cancer: A Multicenter Study.

OBJECTIVE: To investigate the impact of thoracic body composition on outcomes after lobectomy for lung cancer. SUMMARY AND BACKGROUND DATA: Preoperative identification of patients at risk for adverse outcomes permits treatment modification. The impact of body composition on lung resection outcomes has not been investigated in a multicenter setting. METHODS: A total of 958 consecutive patients undergoing lobectomy for lung cancer at 3 centers from 2014 to 2017 were retrospectively analyzed. Muscle and adipose tissue cross-sectional area at the fifth, eighth, and tenth thoracic vertebral body was quantified. Prospectively collected outcomes from a national database were abstracted to characterize the association between sums of muscle and adipose tissue and hospital length of stay (LOS), number of any postoperative complications, and number of respiratory postoperative complications using multivariate regression. A priori determined covariates were forced expiratory volume in 1 second and diffusion capacity of the lungs for carbon monoxide predicted, age, sex, body mass index, race, surgical approach, smoking status, Zubrod and American Society of Anesthesiologists scores. RESULTS: Mean patient age was 67 years, body mass index 27.4 kg/m2 and 65% had stage i disease. Sixty-three percent underwent minimally invasive lobectomy. Median LOS was 4 days and 34% of patients experienced complications. Muscle (using 30 cm2 increments) was an independent predictor of LOS (adjusted coefficient 0.972; P = 0.002), any postoperative complications (odds ratio 0.897; P = 0.007) and postoperative respiratory complications (odds ratio 0.860; P = 0.010). Sarcopenic obesity was also associated with LOS and adverse outcomes. CONCLUSIONS: Body composition on preoperative chest computed tomography is an independent predictor of LOS and postoperative complications after lobectomy for lung cancer.