Impact of a Categorical AI System for Digital Breast Tomosynthesis on Breast Cancer Interpretation by Both General Radiologists and Breast Imaging Specialists.

Purpose To evaluate performance improvements of general radiologists and breast imaging specialists when interpreting a set of diverse digital breast tomosynthesis (DBT) examinations with the aid of a custom-built categorical artificial intelligence (AI) system. Materials and Methods A fully balanced multireader, multicase reader study was conducted to compare the performance of 18 radiologists (nine general radiologists and nine breast imaging specialists) reading 240 retrospectively collected screening DBT mammograms (mean patient age, 59.8 years ± 11.3 [SD]; 100% women), acquired between August 2016 and March 2019, with and without the aid of a custom-built categorical AI system. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity across general radiologists and breast imaging specialists reading with versus without AI were assessed. Reader performance was also analyzed as a function of breast cancer characteristics and patient subgroups. Results Every radiologist demonstrated improved interpretation performance when reading with versus without AI, with an average AUC of 0.93 versus 0.87, demonstrating a difference in AUC of 0.06 (95% CI: 0.04, 0.08; P < .001). Improvement in AUC was observed for both general radiologists (difference of 0.08; P < .001) and breast imaging specialists (difference of 0.04; P < .001) and across all cancer characteristics (lesion type, lesion size, and pathology) and patient subgroups (race and ethnicity, age, and breast density) examined. Conclusion A categorical AI system helped improve overall radiologist interpretation performance of DBT screening mammograms for both general radiologists and breast imaging specialists and across various patient subgroups and breast cancer characteristics. Keywords: Computer-aided Diagnosis, Screening Mammography, Digital Breast Tomosynthesis, Breast Cancer, Screening, Convolutional Neural Network (CNN), Artificial Intelligence Supplemental material is available for this article. © RSNA, 2024.

Erratum: Simultaneous Exercise and Cognitive Training in Virtual Reality Phase 2 Pilot Study: Impact on Brain Health and Cognition in Older Adults.

[This corrects the article DOI: 10.3233/BPL210126.].

Simultaneous Exercise and Cognitive Training in Virtual Reality Phase 2 Pilot Study: Impact on Brain Health and Cognition in Older Adults.

BACKGROUND: Aerobic exercise and environmental enrichment have been shown to enhance brain function. Virtual reality (VR) is a promising method for combining these activities in a meaningful and ecologically valid way. OBJECTIVE: The purpose of this Phase 2 pilot study was to calculate relative change and effect sizes to assess the impact of simultaneous exercise and cognitive training in VR on brain health and cognition in older adults. METHODS: Twelve cognitively normal older adults (64.7±8.8 years old, 8 female) participated in a 12-week intervention, 3 sessions/week for 25-50 minutes/session at 50-80% HRmax. Participants cycled on a custom-built stationary exercise bike while wearing a VR head-mounted display and navigating novel virtual environments to train spatial memory. Brain and cognitive changes were assessed using MRI imaging and a cognitive battery. RESULTS: Medium effect size (ES) improvements in cerebral flow and brain structure were observed. Pulsatility, a measure of peripheral vascular resistance, decreased 10.5% (ES(d) = 0.47). Total grey matter volume increased 0.73% (ES(r) = 0.38), while thickness of the superior parietal lobule, a region associated with spatial orientation, increased 0.44% (ES(r) = 0.30). Visual memory discrimination related to pattern separation showed a large improvement of 68% (ES(η p 2) = 0.43). Cognitive flexibility (Trail Making Test B) (ES(r) = 0.42) and response inhibition (ES(W) = 0.54) showed medium improvements of 14% and 34%, respectively. CONCLUSIONS: Twelve weeks of simultaneous exercise and cognitive training in VR elicits positive changes in brain volume, vascular resistance, memory, and executive function with moderate-to-large effect sizes in our pilot study.

Conducting a VR Clinical Trial in the Era of COVID-19.

The outbreak of severe acute respiratory syndrome coronavirus 2, also known as Coronavirus Disease 2019 (COVID-19) sparked a global public health pandemic that has impacted every aspect of daily life. Medical research was affected, and many clinical trials were halted to minimize COVID-19 transmission risk and spread while the world navigated this novel virus. Here we describe the relaunch of our virtual reality (VR) pilot clinical trial that uses an in-lab brain and body training program to promote brain health in mid-to-late life older adults, in the era of COVID-19. This case series includes five healthy female participants between 51 and 76 years of age, a subset of a larger VR pilot clinical trial that started pre-pandemic. We developed a revised study protocol based on the Center for Disease Control and World Health Organization guidelines to help manage the spread of COVID-19. Since the limited resumption of clinical trials at our institution in August 2020, we successfully completed over 200 in-lab virtual reality training sessions using our revised protocol. During this time, none of the five participants or three study staff reported any COVID-19 symptoms or reported a positive COVID-19 test. More than 40 voluntary COVID-19 tests were completed by our study staff over the last 6 months. All participants rated our safety protocol as very satisfied or extremely satisfied and that they would be very likely or extremely likely to participate in a VR clinical trial during the pandemic. Based on these findings, we suggest that continued VR clinical trial research during the COVID-19 pandemic is achievable and can be safely resumed if specific safety protocols are in place to mitigate the risk of exposure and spread of COVID-19.

Cycling and Spatial Navigation in an Enriched, Immersive 3D Virtual Park Environment: A Feasibility Study in Younger and Older Adults.

BACKGROUND: Cognitive decline is a significant public health concern in older adults. Identifying new ways to maintain cognitive and brain health throughout the lifespan is of utmost importance. Simultaneous exercise and cognitive engagement has been shown to enhance brain function in animal and human studies. Virtual reality (VR) may be a promising approach for conducting simultaneous exercise and cognitive studies. In this study, we evaluated the feasibility of cycling in a cognitively enriched and immersive spatial navigation VR environment in younger and older adults. METHODS: A total of 20 younger (25.9 ± 3.7 years) and 20 older (63.6 ± 5.6 years) adults participated in this study. Participants completed four trials (2 learning and 2 recall) of cycling while wearing a head-mounted device (HMD) and navigating a VR park environment. Questionnaires were administered to assess adverse effects, mood, presence, and physical exertion levels associated with cycling in the VR environment. RESULTS: A total of 4 subjects withdrew from the study due to adverse effects, yielding a 90% completion rate. Simulator sickness levels were enhanced in both age groups with exposure to the VR environment but were within an acceptable range. Exposure to the virtual environment was associated with high arousal and low stress levels, suggesting a state of excitement, and most participants reported enjoyment of the spatial navigation task and VR environment. No association was found between physical exertion levels and simulator sickness levels. CONCLUSION: This study demonstrates that spatial navigation while cycling is feasible and that older adults report similar experiences to younger adults. VR may be a powerful tool for engaging physical and cognitive activity in older adults with acceptable adverse effects and with reports of enjoyment. Future studies are needed to assess the efficacy of a combined exercise and cognitive VR program as an intervention for promoting healthy brain aging, especially in older adults with increased risk of age-related cognitive decline.

Assessing test-retest reliability of phase contrast MRI for measuring cerebrospinal fluid and cerebral blood flow dynamics.

PURPOSE: Pathological states occur when cerebrospinal fluid (CSF) and cerebral blood flow (CBF) dynamics become dysregulated in the brain. Phase-contrast MRI (PC-MRI) is a noninvasive imaging technique that enables quantitative measurements of CSF and CBF flow. While studies have validated PC-MRI as an imaging technique for flow, few studies have evaluated its reliability for CSF and CBF flow parameters commonly associated with neurological disease. The purpose of this study was to evaluate test-retest reliability at the cerebral aqueduct (CA) and C2-C3 area using PC-MRI to assess the feasibility of investigating CSF and CBF flow dynamics. METHODS: This study was performed on 27 cognitively normal young adults (ages 20-35 years). Flow data was acquired on a 3T Siemens Prisma using a 2D cine-PC pulse sequence. Three consecutive flow measurements were acquired at the CA and C2-C3 area. Intraclass correlation coefficient (ICC) and coefficient of variance (CV) were used to evaluate intrarater, inter-rater, and test-retest reliability. RESULTS: Among the 26 flow parameters analyzed, 22 had excellent reliability (ICC > 0.80), including measurements of CSF stroke volume, flush peak, and fill peak, and 4 parameters had good reliability (ICC 0.60-0.79). 16 flow parameters had a mean CV ≤ 10%, 7 had a CV ≤ 15%, and 3 had a CV ≤ 30%. All CSF and CBF flow measurements had excellent inter-rater and intrarater reliability (ICC > 0.80). CONCLUSION: This study shows that CSF and CBF flow can be reliably measured at the CA and C2-C3 area using PC-MRI, making it a promising tool for studying flow dynamics in the central nervous system.