A comprehensive analysis of SOX17 expression by immunohistochemistry in human epithelial tumors, with an emphasis on gynecologic tumors.

OBJECTIVES: The objective of this study was to evaluate SOX17, a transcription factor from the Sry high-mobility group-related box superfamily, as a diagnostic marker to determine site of origin using both whole-tissue sections and tissue microarrays (TMAs). METHODS: SOX17 immunohistochemistry was performed on gynecologic and nongynecologic tissues (N = 1004) using whole-tissue sections and both internally constructed and commercially available TMAs. SOX17 nuclear reactivity was scored as positive or negative on the whole-tissue sections and using the semiquantitative H score method on TMAs. RESULTS: Using both whole-tissue sections and TMAs, SOX17 was positive in 94% (n = 155) of endometrial tumors and 96% (n = 242) of ovarian tumors. All breast cases (n = 241) and vulvar/cervical squamous cell carcinomas (n = 150) were negative. Among 1004 tumors from 20 sites, the only organs with positive tumors were ovary, uterus, and testis. CONCLUSIONS: SOX17 is a sensitive and specific marker for gynecologic origin in the tissues tested and may be a valuable adjunct to PAX8 and other commonly used markers to confirm endometrial or ovarian origin. SOX17 expression is lower in mucinous tumors, endocervical adenocarcinoma, high-grade neuroendocrine tumors, and undifferentiated/dedifferentiated endometrial carcinoma.

Breast Cancer Screening Interval: Effect on Rate of Late-Stage Disease at Diagnosis and Overall Survival.

PURPOSE: Controversy continues regarding the effect of screening mammography on breast cancer outcomes. We evaluated late-stage cancer rate and overall survival (OS) for different screening intervals using a real-world institutional research data mart. METHODS: Patients having both a cancer registry record of new breast cancer diagnosis and prediagnosis screening history between 2004 and 2019 were identified from our institutional research breast data mart. Time interval between the two screening mammograms immediately preceding diagnosis and the time to cancer diagnosis were determined. Screening interval was deemed annual if ≤15 months, biennial if >15 and ≤27 months, intermittent if >27 months, and baseline if only one prediagnosis screen was known. The primary end point was late-stage cancer (TNM stage IIB or worse), and the secondary end point was OS. The association of screening interval and late-stage cancer was analyzed using multivariable logistic regression adjusting for prediagnosis characteristics. Proportional hazards regression was used for survival analysis. Potential lead time was analyzed using survival from a uniform fixed time point. RESULTS: In total, 8,145 patients with breast cancer had prediagnosis screening mammography in the timeframe. The percentage of late-stage cancers diagnosed increased significantly with screening interval with 9%, 14%, and 19% late stages for annual, biennial, and intermittent groups (P < .001), respectively. The trend persisted regardless of age, race, and menopausal status. Biennial and intermittent groups had substantially worse OS than the annual screened group, with relative hazards of 1.42 (95% CI, 1.11 to 1.82) and 2.69 (95% CI, 2.11 to 3.43), respectively, and 1.39 (95% CI, 1.08 to 1.78) and 2.01 (95% CI, 1.58 to 2.55) after adjustment for potential lead time. CONCLUSION: Annual mammographic screening was associated with lower risk of late-stage cancer and better OS across clinical and demographic subgroups. Our study suggests benefit of annual screening for women 40 years and older.

MANCR lncRNA Modulates Cell-Cycle Progression and Metastasis by Cis-Regulation of Nuclear Rho-GEF.

A significant number of the genetic alterations observed in cancer patients lie within nonprotein-coding segments of the genome, including regions coding for long noncoding RNAs (lncRNAs). LncRNAs display aberrant expression in breast cancer (BrCa), but the functional implications of this altered expression remain to be elucidated. By performing transcriptome screen in a triple negative BrCa (TNBC) isogenic 2D and 3D spheroid model, we observed aberrant expression of >1000 lncRNAs during BrCa progression. The chromatin-associated lncRNA MANCR shows elevated expression in metastatic TNBC. MANCR is upregulated in response to cellular stress and modulates DNA repair and cell proliferation. MANCR promotes metastasis as MANCR-depleted cells show reduced cell migration, invasion, and wound healing in vitro, and reduced metastatic lung colonization in xenograft experiments in vivo. Transcriptome analyses reveal that MANCR modulates expression and pre-mRNA splicing of genes, controlling DNA repair and checkpoint response. MANCR promotes the transcription of NET1A, a Rho-GEF that regulates DNA damage checkpoint and metastatic processes in cis, by differential promoter usage. Experiments suggest that MANCR regulates the expression of cancer-associated genes by modulating the association of various transcription factors and RNA-binding proteins. Our results identified the metastasis-promoting activities of MANCR in TNBC by cis-regulation of gene expression.

Inspiring a convergent engineering approach to measure and model the tissue microenvironment.

Understanding the molecular and physical complexity of the tissue microenvironment (TiME) in the context of its spatiotemporal organization has remained an enduring challenge. Recent advances in engineering and data science are now promising the ability to study the structure, functions, and dynamics of the TiME in unprecedented detail; however, many advances still occur in silos that rarely integrate information to study the TiME in its full detail. This review provides an integrative overview of the engineering principles underlying chemical, optical, electrical, mechanical, and computational science to probe, sense, model, and fabricate the TiME. In individual sections, we first summarize the underlying principles, capabilities, and scope of emerging technologies, the breakthrough discoveries enabled by each technology and recent, promising innovations. We provide perspectives on the potential of these advances in answering critical questions about the TiME and its role in various disease and developmental processes. Finally, we present an integrative view that appreciates the major scientific and educational aspects in the study of the TiME.

Capturing cell morphology dynamics with high temporal resolution using single-shot quantitative phase gradient imaging.

Significance: Label-free quantitative phase imaging can potentially measure cellular dynamics with minimal perturbation, motivating efforts to develop faster and more sensitive instrumentation. We characterize fast, single-shot quantitative phase gradient microscopy (ss-QPGM) that simultaneously acquires multiple polarization components required to reconstruct phase images. We integrate a computationally efficient least squares algorithm to provide real-time, video-rate imaging (up to 75 frames / s ). The developed instrument was used to observe changes in cellular morphology and correlate these to molecular measures commonly obtained by staining. Aim: We aim to characterize a fast approach to ss-QPGM and record morphological changes in single-cell phase images. We also correlate these with biochemical changes indicating cell death using concurrently acquired fluorescence images. Approach: Here, we examine nutrient deprivation and anticancer drug-induced cell death in two different breast cell lines, viz., M2 and MCF7. Our approach involves in-line measurements of ss-QPGM and fluorescence imaging of the cells biochemically labeled for viability. Results: We validate the accuracy of the phase measurement using a USAF1951 pattern phase target. The ss-QPGM system resolves 912.3 lp / mm , and our analysis scheme accurately retrieves the phase with a high correlation coefficient ( ∼ 0.99 ), as measured by calibrated sample thicknesses. Analyzing the contrast in phase, we estimate the spatial resolution achievable to be 0.55 µ m for this microscope. ss-QPGM time-lapse live-cell imaging reveals multiple intracellular and morphological changes during biochemically induced cell death. Inferences from co-registered images of quantitative phase and fluorescence suggest the possibility of necrosis, which agrees with previous findings. Conclusions: Label-free ss-QPGM with high-temporal resolution and high spatial fidelity is demonstrated. Its application for monitoring dynamic changes in live cells offers promising prospects.

Spatial molecular profiling of mixed invasive ductal and lobular breast cancers reveals heterogeneity in intrinsic molecular subtypes, oncogenic signatures, and mutations.

Mixed invasive ductal and lobular carcinoma (MDLC) is a rare histologic subtype of breast cancer displaying both E-cadherin positive ductal and E-cadherin negative lobular morphologies within the same tumor, posing challenges with regard to anticipated clinical management. It remains unclear whether these distinct morphologies also have distinct biology and risk of recurrence. Our spatially resolved transcriptomic, genomic, and single-cell profiling revealed clinically significant differences between ductal and lobular tumor regions including distinct intrinsic subtype heterogeneity - e.g., MDLC with triple-negative breast cancer (TNBC) or basal ductal and estrogen receptor positive (ER+) luminal lobular regions, distinct enrichment of cell cycle arrest/senescence and oncogenic (ER and MYC) signatures, genetic and epigenetic CDH1 inactivation in lobular but not ductal regions, and single-cell ductal and lobular subpopulations with unique oncogenic signatures further highlighting intraregional heterogeneity. Altogether, we demonstrated that the intratumoral morphological/histological heterogeneity within MDLC is underpinned by intrinsic subtype and oncogenic heterogeneity which may result in prognostic uncertainty and therapeutic dilemma.

INSTRAS: INfrared Spectroscopic imaging-based TRAnsformers for medical image Segmentation.

Infrared (IR) spectroscopic imaging is of potentially wide use in medical imaging applications due to its ability to capture both chemical and spatial information. This complexity of the data both necessitates using machine intelligence as well as presents an opportunity to harness a high-dimensionality data set that offers far more information than today's manually-interpreted images. While convolutional neural networks (CNNs), including the well-known U-Net model, have demonstrated impressive performance in image segmentation, the inherent locality of convolution limits the effectiveness of these models for encoding IR data, resulting in suboptimal performance. In this work, we propose an INfrared Spectroscopic imaging-based TRAnsformers for medical image Segmentation (INSTRAS). This novel model leverages the strength of the transformer encoders to segment IR breast images effectively. Incorporating skip-connection and transformer encoders, INSTRAS overcomes the issue of pure convolution models, such as the difficulty of capturing long-range dependencies. To evaluate the performance of our model and existing convolutional models, we conducted training on various encoder-decoder models using a breast dataset of IR images. INSTRAS, utilizing 9 spectral bands for segmentation, achieved a remarkable AUC score of 0.9788, underscoring its superior capabilities compared to purely convolutional models. These experimental results attest to INSTRAS's advanced and improved segmentation abilities for IR imaging.

Laparoscopic Total Pancreatectomy After Laparoscopic Whipple.

BACKGROUND: The incidence of a second de novo pancreatic ductal adenocarcinoma (PDAC) among patients with prior cancer has been reported to be 6%.1,2 however, as survival increases through improvements in systemic therapy, this incidence of a de novo PDAC after prior PDAC may become more prevalent.3-8 In this context, a structured and stepwise approach to a total pancreatectomy for a second de novo PDAC after a prior PDAC treated with a pancreaticoduodenectomy is detailed. PATIENTS: We present two similar cases. The first patient was a 71-year-old female with de novo body PDAC, and the second was a 50-year-old female with de novo tail PDAC. To rule out recurrence, immunohistochemical staining as well as the review of biopsies by two experienced pathologists were employed. Both patients had undergone a laparoscopic pancreatoduodenectomy for PDAC 4 and 3 years prior. Each patient received four cycles of neoadjuvant chemotherapy and underwent a safe laparoscopic total pancreatectomy. TECHNIQUE: Prior to surgery, three-dimensional anatomic and port site modeling is performed to optimize the understanding of the spatial relationship between the tumor, blood vessels, and adjacent organs involved. The port site modeling (including pneumoperitoneum simulation) focuses on the optimal port set-up for dissecting the biliopancreatic limb off the portal vein. Following complete mobilization of the biliopancreatic limb, the biliopancreatic limb is staple-divided between the hepatico- and pancreaticojejunostomy. Great care must be taken to avoid accidental staple injury to the hepatic artery or celiac trunk. The remainder of the dissection is akin to a standard distal pancreaticosplenectomy. CONCLUSION: Virtual pancreatectomy modeling facilitates an optimal set-up for the critical step of this case, i.e. dissection of the pancreaticojejunostomy off the portal vein. Early division of the biliopancreatic limb between hepatico- and pancreatojejunostomy is crucial to facilitating the remainder of the dissection. Laparoscopic total pancreatectomy for a de novo PDAC after laparoscopic pancreaticoduodenectomy may become more common as survival of patients with prior PDAC improves over time.

Spatial molecular profiling of mixed invasive ductal-lobular breast cancers reveals heterogeneity in intrinsic molecular subtypes, oncogenic signatures, and mutations.

Mixed invasive ductal and lobular carcinoma (MDLC) is a rare histologic subtype of breast cancer displaying both E-cadherin positive ductal and E-cadherin negative lobular morphologies within the same tumor, posing challenges with regard to anticipated clinical management. It remains unclear whether these distinct morphologies also have distinct biology and risk of recurrence. Our spatially-resolved transcriptomic, genomic, and single-cell profiling revealed clinically significant differences between ductal and lobular tumor regions including distinct intrinsic subtype heterogeneity (e.g., MDLC with TNBC/basal ductal and ER+/luminal lobular regions), distinct enrichment of senescence/dormancy and oncogenic (ER and MYC) signatures, genetic and epigenetic CDH1 inactivation in lobular, but not ductal regions, and single-cell ductal and lobular sub-populations with unique oncogenic signatures further highlighting intra-regional heterogeneity. Altogether, we demonstrated that the intra-tumoral morphological/histological heterogeneity within MDLC is underpinned by intrinsic subtype and oncogenic heterogeneity which may result in prognostic uncertainty and therapeutic dilemma. Significance: MDLC displays both ductal and lobular tumor regions. Our multi-omic profiling approach revealed that these morphologically distinct tumor regions harbor distinct intrinsic subtypes and oncogenic features that may cause prognostic uncertainty and therapeutic dilemma. Thus histopathological/molecular profiling of individual tumor regions may guide clinical decision making and benefit patients with MDLC, particularly in the advanced setting where there is increased reliance on next generation sequencing.

MADR-Net: multi-level attention dilated residual neural network for segmentation of medical images.

Medical image segmentation has made a significant contribution towards delivering affordable healthcare by facilitating the automatic identification of anatomical structures and other regions of interest. Although convolution neural networks have become prominent in the field of medical image segmentation, they suffer from certain limitations. In this study, we present a reliable framework for producing performant outcomes for the segmentation of pathological structures of 2D medical images. Our framework consists of a novel deep learning architecture, called deep multi-level attention dilated residual neural network (MADR-Net), designed to improve the performance of medical image segmentation. MADR-Net uses a U-Net encoder/decoder backbone in combination with multi-level residual blocks and atrous pyramid scene parsing pooling. To improve the segmentation results, channel-spatial attention blocks were added in the skip connection to capture both the global and local features and superseded the bottleneck layer with an ASPP block. Furthermore, we introduce a hybrid loss function that has an excellent convergence property and enhances the performance of the medical image segmentation task. We extensively validated the proposed MADR-Net on four typical yet challenging medical image segmentation tasks: (1) Left ventricle, left atrium, and myocardial wall segmentation from Echocardiogram images in the CAMUS dataset, (2) Skin cancer segmentation from dermoscopy images in ISIC 2017 dataset, (3) Electron microscopy in FIB-SEM dataset, and (4) Fluid attenuated inversion recovery abnormality from MR images in LGG segmentation dataset. The proposed algorithm yielded significant results when compared to state-of-the-art architectures such as U-Net, Residual U-Net, and Attention U-Net. The proposed MADR-Net consistently outperformed the classical U-Net by 5.43%, 3.43%, and 3.92% relative improvement in terms of dice coefficient, respectively, for electron microscopy, dermoscopy, and MRI. The experimental results demonstrate superior performance on single and multi-class datasets and that the proposed MADR-Net can be utilized as a baseline for the assessment of cross-dataset and segmentation tasks.

Modeling the Thermoelastic Sample Response for Subdiffraction Infrared Spectroscopic Imaging.

There is significant and increasing interest in using the photothermal effect to record infrared (IR) absorption spectra localized to volumes that are considerably smaller than the wavelength of excitation, i.e., subdiffraction imaging. As opposed to conventional IR microscopy, in which absorption and scattering of the illuminating light is measured, subdiffraction imaging can be achieved through detection of the sample's thermal response to IR absorption-induced heating. While this relationship has been examined by a variety of coarse-grained models, a generalized analysis of the dependence of temperature and surface deformation arising from an absorber below the surface has not been reported. Here, we present an analytical model to understand a sample's thermoelastic response in photothermal measurements. The model shows important dependence of the ability to record subdiffraction data on modulation frequency of exciting light, limitations imposed by optical sensing, and the potential to discern location of objects ultimately limited by noise and sharpness of the detecting mechanism. This foundational analysis should allow for better modeling, understanding, and harnessing of the relationship between absorption and sample response that underlies IR photothermal measurements.

Label-Free Monitoring of Coculture System Dynamics: Probing Probiotic and Cancer Cell Interactions via Infrared Spectroscopic Imaging.

Evaluating the dynamic interaction of microorganisms and mammalian cells is challenging due to the lack of suitable platforms for examining interspecies interactions in biologically relevant coculture conditions. In this work, we demonstrate the interaction between probiotic bacteria (Lactococcus lactis and Escherichia coli) and A498 human cancer cells in vitro, utilizing a hydrogel-based platform in a label-free manner by infrared spectroscopy. The L. lactis strain recapitulated in the compartment system secretes polypeptide molecules such as nisin, which has been reported to trigger cell apoptosis. We propose a mid-infrared (IR) spectroscopic imaging approach to monitor the variation of biological components utilizing kidney cells (A498) as a model system cocultured with bacteria. We characterized the biochemical composition (i.e., nucleic acids, protein secondary structures, and lipid conformations) label-free using an unbiased measurement. Several IR spectral features, including unsaturated fatty acids, ß-turns in protein, and nucleic acids, were utilized to predict cellular response. These features were then applied to establish a quantitative relationship through a multivariate regression model to predict cellular dynamics in the coculture system to assess the effect of nisin on A498 kidney cancer cells cocultured with bacteria. Overall, our study sheds light on the potential of using IR spectroscopic imaging as a label-free tool to monitor complex microbe-host cell interactions in biological systems. This integration will enable mechanistic studies of interspecies interactions with insights into their underlying physiological processes.

Immunohistochemical Profile of Triple-Negative Breast Cancers: SOX10 and AR Dual Negative Tumors Have Worse Outcomes.

Triple-negative breast cancer (TNBC) refers to an estrogen receptor-negative, progesterone receptor-negative, and HER2-negative breast cancer. Although accepted as a clinically valid category, TNBCs are heterogeneous at the histologic, immunohistochemical, and molecular levels. Gene expression profiling studies have molecularly classified TNBCs into multiple groups, but the prognostic significance is unclear except for a relatively good prognosis for the luminal androgen receptor subtype. Immunohistochemistry (IHC) has been used as a surrogate for basal and luminal subtypes within TNBC, but prognostication of TNBC using IHC is not routinely performed. We aimed to study immunophenotypic correlations in a well-annotated cohort of consecutive TNBCs, excluding postneoadjuvant chemotherapy cases. Tissue microarrays were constructed from a total of 245 TNBC cases. IHC stains were performed and consisted of luminal (AR and INPP4B), basal (SOX10, nestin, CK5, and EGFR), and diagnostic (GCDFP15, mammaglobin, GATA3, and TRPS1) markers. Survival analysis was performed to assess the significance of clinical-pathologic variables including age, histology, grade, lymphovascular invasion, Nottingham prognostic index category, American Joint Committee on Cancer (AJCC) stage, stromal tumor-infiltrating lymphocytes at 10% increment, CD8+ T-cell count, Ki-67 index, PD-L1 status, and chemotherapy along with the results of IHC markers. Apocrine tumors show prominent reactivity for luminal markers and GCDFP15, whereas no special-type carcinomas are often positive for basal markers. TRPS1 is a sensitive marker of breast carcinoma but shows low or no expression in apocrine tumors. High AJCC stage, lack of chemotherapy, and dual SOX10/AR negativity are associated with worse outcomes on both univariable and multivariable analyses. Lymphovascular invasion and higher Nottingham prognostic index category were associated with worse outcomes on univariable but not multivariable analysis. The staining for IHC markers varies based on tumor histology, which may be considered in determining breast origin. Notably, we report that SOX10/AR dual negative status in TNBC is associated with a worse prognosis along with AJCC stage and chemotherapy status.

Osteopontin stabilization and collagen containment slows amorphous calcium phosphate transformation during human aortic valve leaflet calcification.

Calcification of aortic valve leaflets is a growing mortality threat for the 18 million human lives claimed globally each year by heart disease. Extensive research has focused on the cellular and molecular pathophysiology associated with calcification, yet the detailed composition, structure, distribution and etiological history of mineral deposition remains unknown. Here transdisciplinary geology, biology and medicine (GeoBioMed) approaches prove that leaflet calcification is driven by amorphous calcium phosphate (ACP), ACP at the threshold of transformation toward hydroxyapatite (HAP) and cholesterol biomineralization. A paragenetic sequence of events is observed that includes: (1) original formation of unaltered leaflet tissues: (2) individual and coalescing 100's nm- to 1 µm-scale ACP spherules and cholesterol crystals biomineralizing collagen fibers and smooth muscle cell myofilaments; (3) osteopontin coatings that stabilize ACP and collagen containment of nodules preventing exposure to the solution chemistry and water content of pumping blood, which combine to slow transformation to HAP; (4) mm-scale nodule growth via ACP spherule coalescence, diagenetic incorporation of altered collagen and aggregation with other ACP nodules; and (5) leaflet diastole and systole flexure causing nodules to twist, fold their encasing collagen fibers and increase stiffness. These in vivo mechanisms combine to slow leaflet calcification and establish previously unexplored hypotheses for testing novel drug therapies and clinical interventions as viable alternatives to current reliance on surgical/percutaneous valve implants.

Sensitive Discrete Frequency Mid-Infrared Absorption Spectroscopy Using Digitally Referenced Detection.

Broadly tunable mid-infrared (IR) lasers, including quantum cascade lasers (QCL), are an asset for vibrational spectroscopy wherein high-intensity, coherent illumination can target specific spectral bands for rapid, direct chemical detection with microscopic localization. These emerging spectrometers are capable of high measurement throughputs with large detector signals from the high-intensity lasers and fast detection speeds as short as a single laser pulse, challenging the decades old benchmarks of Fourier transform infrared spectroscopy. However, noise in QCL emissions limits the feasible acquisition time for high signal-to-noise ratio (SNR) data. Here, we present an implementation that is broadly compatible with many laser-based spectrometer and microscope designs to address these limitations by leveraging high-speed digitizers and dual detectors to digitally reference each pulse individually. Digitally referenced detection (DRD) is shown to improve measurement sensitivity, with broad spectral indifference, regardless of imbalance due to dissimilarities among system designs or component manufacturers. We incorporated DRD into existing instruments and demonstrated its generalizability: a spectrometer with a 10-fold reduction in spectral noise, a microscope with reduced pixel dwell times to as low as 1 pulse while maintaining SNR normally achieved when operating 8-fold slower, and finally, a spectrometer to measure vibrational circular dichroism (VCD) with a ∼ 4-fold reduction in scan times. The approach not only proves versatile and effective but can also be tailored for specific applications with minimal hardware changes, positioning it as a simple and promising module for spectrometer designs using lasers.

Spectroelectrochemical determination of thiolate self-assembled monolayer adsorptive stability in aqueous and non-aqueous electrolytes.

Self-assembled monolayers (SAM) are ubiquitous in studies of modified electrodes for sensing, electrocatalysis, and environmental and energy applications. However, determining their adsorptive stability is crucial to ensure robust experiments. In this work, the stable potential window (SPW) in which a SAM-covered electrode can function without inducing SAM desorption was determined for aromatic SAMs on gold electrodes in aqueous and non-aqueous solvents. The SPWs were determined by employing cyclic voltammetry, attenuated total reflectance surface-enhanced infrared absorption spectroscopy (ATR-SEIRAS), and surface plasmon resonance (SPR). The electrochemical and spectroscopic findings concluded that all the aromatic SAMs used displayed similar trends and SPWs. In aqueous systems, the SPW lies between the reductive desorption and oxidative desorption, with pH being the decisive factor affecting the range of the SPW, with the widest SPW observed at pH 1. In the non-aqueous electrolytes, the desorption of SAMs was observed to be slow and progressive. The polarity of the solvent was the main factor in determining the SPW. The lower the polarity of the solvent, the larger the SPW, with 1-butanol displaying the widest SPW. This work showcases the power of spectroelectrochemical analysis and provides ample future directions for the use of non-polar solvents to increase SAM stability in electrochemical applications.

Organizational Breast Cancer Data Mart: A Solution for Assessing Outcomes of Imaging and Treatment.

PURPOSE: In the United States, a comprehensive national breast cancer registry (CR) does not exist. Thus, care and coverage decisions are based on data from population subsets, other countries, or models. We report a prototype real-world research data mart to assess mortality, morbidity, and costs for breast cancer diagnosis and treatment. METHODS: With institutional review board approval and Health Insurance Portability and Accountability Act (HIPPA) compliance, a multidisciplinary clinical and research data warehouse (RDW) expert group curated demographic, risk, imaging, pathology, treatment, and outcome data from the electronic health records (EHR), radiology (RIS), and CR for patients having breast imaging and/or a diagnosis of breast cancer in our institution from January 1, 2004, to December 31, 2020. Domains were defined by prebuilt views to extract data denormalized according to requirements from the existing RDW using an export, transform, load pattern. Data dictionaries were included. Structured query language was used for data cleaning. RESULTS: Five-hundred eighty-nine elements (EHR 311, RIS 211, and CR 67) were mapped to 27 domains; all, except one containing CR elements, had cancer and noncancer cohort views, resulting in a total of 53 views (average 12 elements/view; range, 4-67). EHR and RIS queries returned 497,218 patients with 2,967,364 imaging examinations and associated visit details. Cancer biology, treatment, and outcome details for 15,619 breast cancer cases were imported from the CR of our primary breast care facility for this prototype mart. CONCLUSION: Institutional real-world data marts enable comprehensive understanding of care outcomes within an organization. As clinical data sources become increasingly structured, such marts may be an important source for future interinstitution analysis and potentially an opportunity to create robust real-world results that could be used to support evidence-based national policy and care decisions for breast cancer.

Rapid and Label-Free Histopathology of Oral Lesions Using Deep Learning Applied to Optical and Infrared Spectroscopic Imaging Data.

Oral potentially malignant disorders (OPMDs) are precursors to over 80% of oral cancers. Hematoxylin and eosin (H&E) staining, followed by pathologist interpretation of tissue and cellular morphology, is the current gold standard for diagnosis. However, this method is qualitative, can result in errors during the multi-step diagnostic process, and results may have significant inter-observer variability. Chemical imaging (CI) offers a promising alternative, wherein label-free imaging is used to record both the morphology and the composition of tissue and artificial intelligence (AI) is used to objectively assign histologic information. Here, we employ quantum cascade laser (QCL)-based discrete frequency infrared (DFIR) chemical imaging to record data from oral tissues. In this proof-of-concept study, we focused on achieving tissue segmentation into three classes (connective tissue, dysplastic epithelium, and normal epithelium) using a convolutional neural network (CNN) applied to three bands of label-free DFIR data with paired darkfield visible imaging. Using pathologist-annotated H&E images as the ground truth, we demonstrate results that are 94.5% accurate with the ground truth using combined information from IR and darkfield microscopy in a deep learning framework. This chemical-imaging-based workflow for OPMD classification has the potential to enhance the efficiency and accuracy of clinical oral precancer diagnosis.