RESUMO
Diffusion tensor imaging (DTI) can provide unique contrast and insight into microstructural changes with age or disease of the hippocampus, although it is difficult to measure the hippocampus because of its comparatively small size, location, and shape. This has been markedly improved by the advent of a clinically feasible 1-mm isotropic resolution 6-min DTI protocol at 3 T of the hippocampus with limited brain coverage of 20 axial-oblique slices aligned along its long axis. However, manual segmentation is too laborious for large population studies, and it cannot be automatically segmented directly on the diffusion images using traditional T1 or T2 image-based methods because of the limited brain coverage and different contrast. An automatic method is proposed here that segments the hippocampus directly on high-resolution diffusion images based on an extension of well-known deep learning architectures like UNet and UNet++ by including additional dense residual connections. The method was trained on 100 healthy participants with previously performed manual segmentation on the 1-mm DTI, then evaluated on typical healthy participants (n = 53), yielding an excellent voxel overlap with a Dice score of ~ 0.90 with manual segmentation; notably, this was comparable with the inter-rater reliability of manually delineating the hippocampus on diffusion magnetic resonance imaging (MRI) (Dice score of 0.86). This method also generalized to a different DTI protocol with 36% fewer acquisitions. It was further validated by showing similar age trajectories of volumes, fractional anisotropy, and mean diffusivity from manual segmentations in one cohort (n = 153, age 5-74 years) with automatic segmentations from a second cohort without manual segmentations (n = 354, age 5-90 years). Automated high-resolution diffusion MRI segmentation of the hippocampus will facilitate large cohort analyses and, in future research, needs to be evaluated on patient groups.
RESUMO
Iatrogenic damage to the cardiac conduction system (CCS) remains a significant risk during congenital heart surgery. Current surgical best practice involves using superficial anatomical landmarks to locate and avoid damaging the CCS. Prior work indicates inherent variability in the anatomy of the CCS and supporting tissues. This study introduces high-resolution, 3D models of the CCS in normal pediatric human hearts to evaluate variability in the nodes and surrounding structures. Human pediatric hearts were obtained with an average donor age of 2.7 days. A pipeline was developed to excise, section, stain, and image atrioventricular (AVN) and sinus nodal (SN) tissue regions. A convolutional neural network was trained to enable precise multi-class segmentation of whole-slide images, which were subsequently used to generate high- resolution 3D tissue models. Nodal tissue region models were created. All models (10 AVN, 8 SN) contain tissue composition of neural tissue, vasculature, and nodal tissues at micrometer resolution. We describe novel nodal anatomical variations. We found that the depth of the His bundle in females was on average 304 µm shallower than those of male patients. These models provide surgeons with insight into the heterogeneity of the nodal regions and the intricate relationships between the CCS and surrounding structures.
Assuntos
Nó Atrioventricular , Imageamento Tridimensional , Humanos , Feminino , Masculino , Recém-Nascido , Nó Atrioventricular/anatomia & histologia , Modelos Cardiovasculares , Nó Sinoatrial/anatomia & histologia , Fascículo Atrioventricular/fisiopatologia , Redes Neurais de Computação , Fatores Sexuais , Fatores Etários , Sistema de Condução Cardíaco/fisiopatologiaRESUMO
Background: Current health behavior recommendations for skin cancer prevention, treatment, and survivorship are the same for survivors of other cancers; they include eating a healthy diet, being physically active, maintaining a healthy weight, and minimizing ultraviolet (U.V.) exposure. Few interventions exist to support health behaviors beyond U.V. exposure. We adapted Harvest for Health, a home-based mentored gardening intervention for cancer survivors, for implementation in Arizona as a community-based intervention. Methods: Stakeholder-informed adaptations for Harvest for Health Together Arizona (H4H2-AZ) included updating intervention materials to be relevant to the arid desert environment, emphasizing the importance of sun safety in cancer survivorship, and shifting from a home-based to a community-based delivery model. Participants will be enrolled in cohorts aligned with growing seasons (e.g., spring, monsoon, fall) and matched to an individual 30 ft2 community garden plot for two growing seasons (6 months). Original intervention components retained are: 1) Master Gardeners deliver the intervention providing one-to-one mentorship and 2) gardening materials and supplies provided. This pilot six-month single-arm intervention will determine feasibility, acceptability, and appropriateness of an evidence-based adapted mentored community gardening intervention for survivors of skin cancer as primary outcomes. Secondary outcomes are to explore the effects on cancer preventive health behaviors and health-related quality of life. Discussion: This pilot single-arm intervention will determine feasibility, acceptability, and appropriateness of an evidence-based adapted mentored community gardening intervention for survivors of skin cancer. If successful, the intervention could be widely implemented throughout existing Master Gardener programs and community garden networks for survivors of other cancers. Trial registration: ClinicalTrials.gov identifier: NCT05648604. Trial registered on December 13, 2022.
RESUMO
OBJECTIVES: To describe mismatch repair (MMR) and microsatellite instability (MSI) testing practices in laboratories using the College of American Pathologists (CAP) MSI/MMR proficiency testing programs prior to the 2022 publication of the MSI/MMR practice guidelines copublished by CAP and the Association of Molecular Pathology (AMP). METHODS: Data from supplemental questionnaires provided with the 2020-B MSI/MMR programs to 542 laboratories across different practice settings were reviewed. Questionnaires contained 21 questions regarding the type of testing performed, specimen/tumor types used for testing, and clinical practices for checkpoint blockade therapy. RESULTS: Domestic laboratories test for MSI/MMR more often than international laboratories (P = .04) and academic hospitals/medical centers test more frequently than nonhospital sites/clinics (P = .03). The most commonly used testing modality is immunohistochemistry, followed by polymerase chain reaction, then next-generation sequencing. Most laboratories (72.6%; 347/478) reported awareness of the use of immune checkpoint inhibitor therapy for patients with high MSI or MMR-deficient results. CONCLUSIONS: The results demonstrate the state of MMR and MSI testing in laboratories prior to the publication of the CAP/AMP best practice guidelines, highlighting differences between various laboratory types. The findings indicate the importance of consensus guidelines and provide a baseline for comparison after their implementation.
RESUMO
BACKGROUND: Pathologic antibody mediated rejection (pAMR) remains a major driver of graft failure in cardiac transplant patients. The endomyocardial biopsy remains the primary diagnostic tool but presents with challenges, particularly in distinguishing the histologic component (pAMR-H) defined by 1) intravascular macrophage accumulation in capillaries and 2) activated endothelial cells that expand the cytoplasm to narrow or occlude the vascular lumen. Frequently, pAMR-H is difficult to distinguish from acute cellular rejection (ACR) and healing injury. With the advent of digital slide scanning and advances in machine deep learning, artificial intelligence technology is widely under investigation in the areas of oncologic pathology, but in its infancy in transplant pathology. For the first time, we determined if a machine learning algorithm could distinguish pAMR-H from normal myocardium, healing injury and ACR. MATERIALS AND METHODS: A total of 4,212 annotations (1,053 regions of normal, 1,053 pAMR-H, 1,053 healing injury and 1,053 ACR) were completed from 300 hematoxylin and eosin slides scanned using a Leica Aperio GT450 digital whole slide scanner at 40X magnification. All regions of pAMR-H were annotated from patients confirmed with a previous diagnosis of pAMR2 (>50% positive C4d immunofluorescence and/or >10% CD68 positive intravascular macrophages). Annotations were imported into a Python 3.7 development environment using the OpenSlide™ package and a convolutional neural network approach utilizing transfer learning was performed. RESULTS: The machine learning algorithm showed 98% overall validation accuracy and pAMR-H was correctly distinguished from specific categories with the following accuracies: normal myocardium (99.2%), healing injury (99.5%) and ACR (99.5%). CONCLUSION: Our novel deep learning algorithm can reach acceptable, and possibly surpass, performance of current diagnostic standards of identifying pAMR-H. Such a tool may serve as an adjunct diagnostic aid for improving the pathologist's accuracy and reproducibility, especially in difficult cases with high inter-observer variability. This is one of the first studies that provides evidence that an artificial intelligence machine learning algorithm can be trained and validated to diagnose pAMR-H in cardiac transplant patients. Ongoing studies include multi-institutional verification testing to ensure generalizability.