Identification of Spatial Proteomic Signatures of Colon Tumor Metastasis: A Digital Spatial Profiling Approach.

Over 150,000 Americans are diagnosed with colorectal cancer (CRC) every year, and annually >50,000 individuals are estimated to die of CRC, necessitating improvements in screening, prognostication, disease management, and therapeutic options. CRC tumors are removed en bloc with surrounding vasculature and lymphatics. Examination of regional lymph nodes at the time of surgical resection is essential for prognostication. Developing alternative approaches to indirectly assess recurrence risk would have utility in cases where lymph node yield is incomplete or inadequate. Spatially dependent, immune cell-specific (eg, tumor-infiltrating lymphocytes), proteomic, and transcriptomic expression patterns inside and around the tumor-the tumor immune microenvironment-can predict nodal/distant metastasis and probe the coordinated immune response from the primary tumor site. The comprehensive characterization of tumor-infiltrating lymphocytes and other immune infiltrates is possible using highly multiplexed spatial omics technologies, such as the GeoMX Digital Spatial Profiler. In this study, machine learning and differential co-expression analyses helped identify biomarkers from Digital Spatial Profiler-assayed protein expression patterns inside, at the invasive margin, and away from the tumor, associated with extracellular matrix remodeling (eg, granzyme B and fibronectin), immune suppression (eg, forkhead box P3), exhaustion and cytotoxicity (eg, CD8), Programmed death ligand 1-expressing dendritic cells, and neutrophil proliferation, among other concomitant alterations. Further investigation of these biomarkers may reveal independent risk factors of CRC metastasis that can be formulated into low-cost, widely available assays.

A deep learning algorithm to detect cutaneous squamous cell carcinoma on frozen sections in Mohs micrographic surgery: A retrospective assessment.

Intraoperative margin analysis is crucial for the successful removal of cutaneous squamous cell carcinomas (cSCC). Artificial intelligence technologies (AI) have previously demonstrated potential for facilitating rapid and complete tumour removal using intraoperative margin assessment for basal cell carcinoma. However, the varied morphologies of cSCC present challenges for AI margin assessment. The aim of this study was to develop and evaluate the accuracy of an AI algorithm for real-time histologic margin analysis of cSCC. To do this, a retrospective cohort study was conducted using frozen cSCC section slides. These slides were scanned and annotated, delineating benign tissue structures, inflammation and tumour to develop an AI algorithm for real-time margin analysis. A convolutional neural network workflow was used to extract histomorphological features predictive of cSCC. This algorithm demonstrated proof of concept for identifying cSCC with high accuracy, highlighting the potential for integration of AI into the surgical workflow. Incorporation of AI algorithms may improve efficiency and completeness of real-time margin assessment for cSCC removal, particularly in cases of moderately and poorly differentiated tumours/neoplasms. Further algorithmic improvement incorporating surrounding tissue context is necessary to remain sensitive to the unique epidermal landscape of well-differentiated tumours, and to map tumours to their original anatomical position/orientation.

Video-Based Deep Learning to Detect Dyssynergic Defecation with 3D High-Definition Anorectal Manometry.

BACKGROUND: We developed a deep learning algorithm to evaluate defecatory patterns to identify dyssynergic defecation using 3-dimensional high definition anal manometry (3D-HDAM). AIMS: We developed a 3D-HDAM deep learning algorithm to evaluate for dyssynergia. METHODS: Spatial-temporal data were extracted from consecutive 3D-HDAM studies performed between 2018 and 2020 at Dartmouth-Hitchcock Health. The technical procedure and gold standard definition of dyssynergia were based on the London consensus, adapted to the needs of 3D-HDAM technology. Three machine learning models were generated: (1) traditional machine learning informed by conventional anorectal function metrics, (2) deep learning, and (3) a hybrid approach. Diagnostic accuracy was evaluated using bootstrap sampling to calculate area-under-the-curve (AUC). To evaluate overfitting, models were validated by adding 502 simulated defecation maneuvers with diagnostic ambiguity. RESULTS: 302 3D-HDAM studies representing 1208 simulated defecation maneuvers were included (average age 55.2 years; 80.5% women). The deep learning model had comparable diagnostic accuracy [AUC 0.91 (95% confidence interval 0.89-0.93)] to traditional [AUC 0.93(0.92-0.95)] and hybrid [AUC 0.96(0.94-0.97)] predictive models in training cohorts. However, the deep learning model handled ambiguous tests more cautiously than other models; the deep learning model was more likely to designate an ambiguous test as inconclusive [odds ratio 4.21(2.78-6.38)] versus traditional/hybrid approaches. CONCLUSIONS: Deep learning is capable of considering complex spatial-temporal information on 3D-HDAM technology. Future studies are needed to evaluate the clinical context of these preliminary findings.

Passive order auditing associated with reductions in red blood cell utilization: National blood shortage experience.

BACKGROUND: Decreased blood collection during the Coronavirus Disease 2019 (COVID-19) pandemic resulted in long-term red blood cell (RBC) shortages in the United States. In an effort to conserve RBCs, the existing passive alert system for auditing inpatient transfusions was modified to activate at a lower hemoglobin threshold (6.5 g/dL instead of 7.0 g/dL for stable, nonbleeding inpatients) during a 9-month shortage at an academic medical center. Hemoglobin levels prior to RBC transfusions were compared for inpatients receiving RBC transfusions to determine whether RBC utilization changed during the intervention. STUDY DESIGN AND METHODS: This retrospective study compared the number of single-unit RBC transfusions and hemoglobin levels prior to RBC transfusion among inpatients during the 9 months of the intervention (Period 2, 06/01/2021-2/28/2022) to the same period of the previous year (Period 1, 06/01/2020-2/28/2021). RESULTS: Overall full unit RBC transfusions to inpatients decreased by 15% from 5182 to 4421. Of all transfusions, 50.3% and 49.8% were single-unit RBC transfusions in Period 1 and Period 2, respectively. The incidence rate difference and incidence rate ratio of single RBC units transfused per 1000 patient days were significantly decreased (p = 0.0007). The average pre-transfusion hemoglobin level significantly decreased from 7.18 g/dL to 7.05 g/dL (p = 0.0002), largely due to significant decreases in hemoglobin transfusion triggers for adult inpatient ward transfusions. DISCUSSION: Modification of the passive alert system was associated with significantly decreased RBC utilization during a long-term RBC shortage. Modification of transfusion criteria recommended by passive alerts may be a feasible option to decrease RBC utilization at centers during long-term RBC shortages.

A large-scale internal validation study of unsupervised virtual trichrome staining technologies on nonalcoholic steatohepatitis liver biopsies.

Non-alcoholic steatohepatitis (NASH) is a fatty liver disease characterized by accumulation of fat in hepatocytes with concurrent inflammation and is associated with morbidity, cirrhosis and liver failure. After extraction of a liver core biopsy, tissue sections are stained with hematoxylin and eosin (H&E) to grade NASH activity, and stained with trichrome to stage fibrosis. Methods to computationally transform one stain into another on digital whole slide images (WSI) can lessen the need for additional physical staining besides H&E, reducing personnel, equipment, and time costs. Generative adversarial networks (GAN) have shown promise for virtual staining of tissue. We conducted a large-scale validation study of the viability of GANs for H&E to trichrome conversion on WSI (n = 574). Pathologists were largely unable to distinguish real images from virtual/synthetic images given a set of twelve Turing Tests. We report high correlation between staging of real and virtual stains ([Formula: see text]; 95% CI: 0.84-0.88). Stages assigned to both virtual and real stains correlated similarly with a number of clinical biomarkers and progression to End Stage Liver Disease (Hazard Ratio HR = 2.06, 95% CI: 1.36-3.12, p < 0.001 for real stains; HR = 2.02, 95% CI: 1.40-2.92, p < 0.001 for virtual stains). Our results demonstrate that virtual trichrome technologies may offer a software solution that can be employed in the clinical setting as a diagnostic decision aid.

MethylNet: an automated and modular deep learning approach for DNA methylation analysis.

BACKGROUND: DNA methylation (DNAm) is an epigenetic regulator of gene expression programs that can be altered by environmental exposures, aging, and in pathogenesis. Traditional analyses that associate DNAm alterations with phenotypes suffer from multiple hypothesis testing and multi-collinearity due to the high-dimensional, continuous, interacting and non-linear nature of the data. Deep learning analyses have shown much promise to study disease heterogeneity. DNAm deep learning approaches have not yet been formalized into user-friendly frameworks for execution, training, and interpreting models. Here, we describe MethylNet, a DNAm deep learning method that can construct embeddings, make predictions, generate new data, and uncover unknown heterogeneity with minimal user supervision. RESULTS: The results of our experiments indicate that MethylNet can study cellular differences, grasp higher order information of cancer sub-types, estimate age and capture factors associated with smoking in concordance with known differences. CONCLUSION: The ability of MethylNet to capture nonlinear interactions presents an opportunity for further study of unknown disease, cellular heterogeneity and aging processes.

PyMethylProcess-convenient high-throughput preprocessing workflow for DNA methylation data.

SUMMARY: Performing highly parallelized preprocessing of methylation array data using Python can accelerate data preparation for downstream methylation analyses, including large scale production-ready machine learning pipelines. We present a highly reproducible, scalable pipeline (PyMethylProcess) that can be quickly set-up and deployed through Docker and PIP. AVAILABILITY AND IMPLEMENTATION: Project Home Page: https://github.com/Christensen-Lab-Dartmouth/PyMethylProcess. Available on PyPI (pymethylprocess), Docker (joshualevy44/pymethylprocess). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Don't dismiss logistic regression: the case for sensible extraction of interactions in the era of machine learning.

BACKGROUND: Machine learning approaches have become increasingly popular modeling techniques, relying on data-driven heuristics to arrive at its solutions. Recent comparisons between these algorithms and traditional statistical modeling techniques have largely ignored the superiority gained by the former approaches due to involvement of model-building search algorithms. This has led to alignment of statistical and machine learning approaches with different types of problems and the under-development of procedures that combine their attributes. In this context, we hoped to understand the domains of applicability for each approach and to identify areas where a marriage between the two approaches is warranted. We then sought to develop a hybrid statistical-machine learning procedure with the best attributes of each. METHODS: We present three simple examples to illustrate when to use each modeling approach and posit a general framework for combining them into an enhanced logistic regression model building procedure that aids interpretation. We study 556 benchmark machine learning datasets to uncover when machine learning techniques outperformed rudimentary logistic regression models and so are potentially well-equipped to enhance them. We illustrate a software package, InteractionTransformer, which embeds logistic regression with advanced model building capacity by using machine learning algorithms to extract candidate interaction features from a random forest model for inclusion in the model. Finally, we apply our enhanced logistic regression analysis to two real-word biomedical examples, one where predictors vary linearly with the outcome and another with extensive second-order interactions. RESULTS: Preliminary statistical analysis demonstrated that across 556 benchmark datasets, the random forest approach significantly outperformed the logistic regression approach. We found a statistically significant increase in predictive performance when using hybrid procedures and greater clarity in the association with the outcome of terms acquired compared to directly interpreting the random forest output. CONCLUSIONS: When a random forest model is closer to the true model, hybrid statistical-machine learning procedures can substantially enhance the performance of statistical procedures in an automated manner while preserving easy interpretation of the results. Such hybrid methods may help facilitate widespread adoption of machine learning techniques in the biomedical setting.

PolyCRACKER, a robust method for the unsupervised partitioning of polyploid subgenomes by signatures of repetitive DNA evolution.

BACKGROUND: Our understanding of polyploid genomes is limited by our inability to definitively assign sequences to a specific subgenome without extensive prior knowledge like high resolution genetic maps or genome sequences of diploid progenitors. In theory, existing methods for assigning sequences to individual species from metagenome samples could be used to separate subgenomes in polyploid organisms, however, these methods rely on differences in coarse genome properties like GC content or sequences from related species. Thus, these approaches do not work for subgenomes where gross features are indistinguishable and related genomes are lacking. Here we describe a method that uses rapidly evolving repetitive DNA to circumvent these limitations. RESULTS: By using short, repetitive, DNA sequences as species-specific signals we separated closely related genomes from test datasets and subgenomes from two polyploid plants, tobacco and wheat, without any prior knowledge. CONCLUSION: This approach is ideal for separating the subgenomes of polyploid species with unsequenced or unknown progenitor genomes.

Maternal glucose levels and late pregnancy circulating extracellular vesicle and particle miRNAs in the MADRES pregnancy cohort.

Maternal hyperglycemia during pregnancy adversely affects maternal and child outcomes. While mechanisms are not fully understood, maternal circulating miRNAs may play a role. We examined whether continuous glucose levels and hyperglycemia subtypes (gestational diabetes, type 2 diabetes, and glucose intolerance) were associated with circulating miRNAs during late pregnancy. Seven miRNAs (hsa-miR-107, hsa-let-7b-5p, hsa-miR-126-3p, hsa-miR-181a-5p, hsa-miR-374a-5p, hsa-miR-382-5p, and hsa-miR-337-5p) were associated (p < 0.05) with either hyperglycemia or continuous glucose levels prior to multiple testing correction. These miRNAs target genes involved in pathways relevant to maternal and child health, including insulin signaling, placental development, energy balance, and appetite regulation.

Spatial Omics Driven Crossmodal Pretraining Applied to Graph-based Deep Learning for Cancer Pathology Analysis.

Graph-based deep learning has shown great promise in cancer histopathology image analysis by contextualizing complex morphology and structure across whole slide images to make high quality downstream outcome predictions (ex: prognostication). These methods rely on informative representations (i.e., embeddings) of image patches comprising larger slides, which are used as node attributes in slide graphs. Spatial omics data, including spatial transcriptomics, is a novel paradigm offering a wealth of detailed information. Pairing this data with corresponding histological imaging localized at 50-micron resolution, may facilitate the development of algorithms which better appreciate the morphological and molecular underpinnings of carcinogenesis. Here, we explore the utility of leveraging spatial transcriptomics data with a contrastive crossmodal pretraining mechanism to generate deep learning models that can extract molecular and histological information for graph-based learning tasks. Performance on cancer staging, lymph node metastasis prediction, survival prediction, and tissue clustering analyses indicate that the proposed methods bring improvement to graph based deep learning models for histopathological slides compared to leveraging histological information from existing schemes, demonstrating the promise of mining spatial omics data to enhance deep learning for pathology workflows.

Intraoperative margin assessment for basal cell carcinoma with deep learning and histologic tumor mapping to surgical site.

Successful treatment of solid cancers relies on complete surgical excision of the tumor either for definitive treatment or before adjuvant therapy. Intraoperative and postoperative radial sectioning, the most common form of margin assessment, can lead to incomplete excision and increase the risk of recurrence and repeat procedures. Mohs Micrographic Surgery is associated with complete removal of basal cell and squamous cell carcinoma through real-time margin assessment of 100% of the peripheral and deep margins. Real-time assessment in many tumor types is constrained by tissue size, complexity, and specimen processing / assessment time during general anesthesia. We developed an artificial intelligence platform to reduce the tissue preprocessing and histological assessment time through automated grossing recommendations, mapping and orientation of tumor to the surgical specimen. Using basal cell carcinoma as a model system, results demonstrate that this approach can address surgical laboratory efficiency bottlenecks for rapid and complete intraoperative margin assessment.

Potential to Enhance Large Scale Molecular Assessments of Skin Photoaging through Virtual Inference of Spatial Transcriptomics from Routine Staining.

The advent of spatial transcriptomics technologies has heralded a renaissance in research to advance our understanding of the spatial cellular and transcriptional heterogeneity within tissues. Spatial transcriptomics allows investigation of the interplay between cells, molecular pathways, and the surrounding tissue architecture and can help elucidate developmental trajectories, disease pathogenesis, and various niches in the tumor microenvironment. Photoaging is the histological and molecular skin damage resulting from chronic/acute sun exposure and is a major risk factor for skin cancer. Spatial transcriptomics technologies hold promise for improving the reliability of evaluating photoaging and developing new therapeutics. Challenges to current methods include limited focus on dermal elastosis variations and reliance on self-reported measures, which can introduce subjectivity and inconsistency. Spatial transcriptomics offers an opportunity to assess photoaging objectively and reproducibly in studies of carcinogenesis and discern the effectiveness of therapies that intervene in photoaging and preventing cancer. Evaluation of distinct histological architectures using highly-multiplexed spatial technologies can identify specific cell lineages that have been understudied due to their location beyond the depth of UV penetration. However, the cost and interpatient variability using state-of-the-art assays such as the 10x Genomics Spatial Transcriptomics assays limits the scope and scale of large-scale molecular epidemiologic studies. Here, we investigate the inference of spatial transcriptomics information from routine hematoxylin and eosin-stained (H&E) tissue slides. We employed the Visium CytAssist spatial transcriptomics assay to analyze over 18,000 genes at a 50-micron resolution for four patients from a cohort of 261 skin specimens collected adjacent to surgical resection sites for basal cell and squamous cell keratinocyte tumors. The spatial transcriptomics data was co-registered with 40x resolution whole slide imaging (WSI) information. We developed machine learning models that achieved a macro-averaged median AUC and F1 score of 0.80 and 0.61 and Spearman coefficient of 0.60 in inferring transcriptomic profiles across the slides, and accurately captured biological pathways across various tissue architectures.

GRANDPA: GeneRAtive network sampling using degree and property augmentation applied to the analysis of partially confidential healthcare networks.

Protecting medical privacy can create obstacles in the analysis and distribution of healthcare graphs and statistical inferences accompanying them. We pose a graph simulation model which generates networks using degree and property augmentation and provide a flexible R package that allows users to create graphs that preserve vertex attribute relationships and approximating the retention of topological properties observed in the original graph (e.g., community structure). We illustrate our proposed algorithm using a case study based on Zachary's karate network and a patient-sharing graph generated from Medicare claims data in 2019. In both cases, we find that community structure is preserved, and normalized root mean square error between cumulative distributions of the degrees across the generated and the original graphs is low (0.0508 and 0.0514 respectively).

HiTAIC: hierarchical tumor artificial intelligence classifier traces tissue of origin and tumor type in primary and metastasized tumors using DNA methylation.

Human cancers are heterogenous by their cell composition and origination site. Cancer metastasis generates the conundrum of the unknown origin of migrated tumor cells. Tracing tissue of origin and tumor type in primary and metastasized cancer is vital for clinical significance. DNA methylation alterations play a crucial role in carcinogenesis and mark cell fate differentiation, thus can be used to trace tumor tissue of origin. In this study, we employed a novel tumor-type-specific hierarchical model using genome-scale DNA methylation data to develop a multilayer perceptron model, HiTAIC, to trace tissue of origin and tumor type in 27 cancers from 23 tissue sites in data from 7735 tumors with high resolution, accuracy, and specificity. In tracing primary cancer origin, HiTAIC accuracy was 99% in the test set and 93% in the external validation data set. Metastatic cancers were identified with a 96% accuracy in the external data set. HiTAIC is a user-friendly web-based application through https://sites.dartmouth.edu/salaslabhitaic/. In conclusion, we developed HiTAIC, a DNA methylation-based algorithm, to trace tumor tissue of origin in primary and metastasized cancers. The high accuracy and resolution of tumor tracing using HiTAIC holds promise for clinical assistance in identifying cancer of unknown origin.

Assessment of emerging pretraining strategies in interpretable multimodal deep learning for cancer prognostication.

BACKGROUND: Deep learning models can infer cancer patient prognosis from molecular and anatomic pathology information. Recent studies that leveraged information from complementary multimodal data improved prognostication, further illustrating the potential utility of such methods. However, current approaches: 1) do not comprehensively leverage biological and histomorphological relationships and 2) make use of emerging strategies to "pretrain" models (i.e., train models on a slightly orthogonal dataset/modeling objective) which may aid prognostication by reducing the amount of information required for achieving optimal performance. In addition, model interpretation is crucial for facilitating the clinical adoption of deep learning methods by fostering practitioner understanding and trust in the technology. METHODS: Here, we develop an interpretable multimodal modeling framework that combines DNA methylation, gene expression, and histopathology (i.e., tissue slides) data, and we compare performance of crossmodal pretraining, contrastive learning, and transfer learning versus the standard procedure. RESULTS: Our models outperform the existing state-of-the-art method (average 11.54% C-index increase), and baseline clinically driven models (average 11.7% C-index increase). Model interpretations elucidate consideration of biologically meaningful factors in making prognosis predictions. DISCUSSION: Our results demonstrate that the selection of pretraining strategies is crucial for obtaining highly accurate prognostication models, even more so than devising an innovative model architecture, and further emphasize the all-important role of the tumor microenvironment on disease progression.

Functional Reorganization of the Mesial Frontal Premotor Cortex in Patients With Supplementary Motor Area Seizures.

BACKGROUND: Direct cortical stimulation of the mesial frontal premotor cortex, including the supplementary motor area (SMA), is challenging in humans. Limited access to these brain regions impedes understanding of human premotor cortex functional organization and somatotopy. OBJECTIVE: To test whether seizure onset within the SMA was associated with functional remapping of mesial frontal premotor areas in a cohort of patients with epilepsy who underwent awake brain mapping after implantation of interhemispheric subdural electrodes. METHODS: Stimulation trials from 646 interhemispheric subdural electrodes were analyzed and compared between patients who had seizure onset in the SMA (n = 13) vs patients who had seizure onset outside of the SMA (n = 12). 1:1 matching with replacement between SMA and non-SMA data sets was used to ensure similar spatial distribution of electrodes. Centroids and 95% confidence regions were computed for clustered head, trunk, upper extremity, lower extremity, and vision responses. A generalized linear mixed-effects model was used to test for significant differences in the resulting functional maps. Clinical, radiographic, and histopathologic data were reviewed. RESULTS: After analyzing direct cortical stimulation trials from interhemispheric electrodes, we found significant displacement of the head and trunk responses in SMA compared with non-SMA patients ( P < .01 for both). These differences remained significant after accounting for structural lesions, preexisting motor deficits, and seizure outcome. CONCLUSION: The somatotopy of the mesial frontal premotor regions is significantly altered in patients who have SMA-onset seizures compared with patients who have seizure onset outside of the SMA, suggesting that functional remapping can occur in these brain regions.

Paired-agent imaging as a rapid en face margin screening method in Mohs micrographic surgery.

Background: Mohs micrographic surgery is a procedure used for non-melanoma skin cancers that has 97-99% cure rates largely owing to 100% margin analysis enabled by en face sectioning with real-time, iterative histologic assessment. However, the technique is limited to small and aggressive tumors in high-risk areas because the histopathological preparation and assessment is very time intensive. To address this, paired-agent imaging (PAI) can be used to rapidly screen excised specimens and identify tumor positive margins for guided and more efficient microscopic evaluation. Methods: A mouse xenograft model of human squamous cell carcinoma (n = 8 mice, 13 tumors) underwent PAI. Targeted (ABY-029, anti-epidermal growth factor receptor (EGFR) affibody molecule) and untargeted (IRDye 680LT carboxylate) imaging agents were simultaneously injected 3-4 h prior to surgical tumor resection. Fluorescence imaging was performed on main, unprocessed excised specimens and en face margins (tissue sections tangential to the deep margin surface). Binding potential (BP) - a quantity proportional to receptor concentration - and targeted fluorescence signal were measured for each, and respective mean and maximum values were analyzed to compare diagnostic ability and contrast. The BP and targeted fluorescence of the main specimen and margin samples were also correlated with EGFR immunohistochemistry (IHC). Results: PAI consistently outperformed targeted fluorescence alone in terms of diagnostic ability and contrast-to-variance ratio (CVR). Mean and maximum measures of BP resulted in 100% accuracy, while mean and maximum targeted fluorescence signal offered 97% and 98% accuracy, respectively. Moreover, maximum BP had the greatest average CVR for both main specimen and margin samples (average 1.7 ± 0.4 times improvement over other measures). Fresh tissue margin imaging improved similarity with EGFR IHC volume estimates compared to main specimen imaging in line profile analysis; and margin BP specifically had the strongest concordance (average 3.6 ± 2.2 times improvement over other measures). Conclusions: PAI was able to reliably distinguish tumor from normal tissue in fresh en face margin samples using the single metric of maximum BP. This demonstrated the potential for PAI to act as a highly sensitive screening tool to eliminate the extra time wasted on real-time pathological assessment of low-risk margins.

Use of molecular testing results to analyze the overuse of atypia of undetermined significance in thyroid cytology.

INTRODUCTION: The suggested atypia of undetermined significance (AUS) rate for thyroid fine-needle aspiration biopsies is 10% or less. Prompted by a high institutional AUS rate, we examined using molecular testing results (MTR) as a potential quality metric tool to reduce the AUS rate. We correlated MTR with AUS cytologic findings, surgical pathology follow-up, and individual pathologist AUS rates. MATERIALS AND METHODS: Demographic data, cytologic diagnoses, MTR, and surgical pathology diagnoses were retrospectively obtained. MTR were classified as either positive or negative. AUS rates and MTR proportions were compared among pathologists. The cytomorphologic features of 143 AUS cases were assessed and correlated with MTR. RESULTS: Between 2017 and 2022, 710 of 3247 thyroid fine-needle aspirations were classified as AUS, with a yearly average rate of 22% (range = 19%-26%). AUS cases included: 331 (47%) with architectural atypia; 204 (29%) with oncocytic (Hürthle cell) atypia; 99 (14%) with combined architectural and cytologic atypia; and 76 (10%) with isolated cytologic atypia. Most AUS cases with molecular testing had negative MTR (360/492, 73%). AUS with cytologic atypia had higher positive MTR risk (logarithm of odds ratio = 1.27, 95% credible interval [0.5-2.04], P = 0.001). The average positive MTR rate by pathologist was 21.5% (range 0%-35%); higher positive MTR rates had better correlation with subsequent neoplastic/malignant histologic diagnoses. The MTR sensitivity for malignant disease was 89% and the negative predictive value was 91%. CONCLUSIONS: MTR analysis reveals the importance of cytologic atypia as a determinant of malignancy risk in AUS cases. Periodic analysis of MTR data alongside individual pathologist AUS rates can help refine diagnostic criteria and potentially reduce AUS overuse.

The Overlooked Role of Specimen Preparation in Bolstering Deep Learning-Enhanced Spatial Transcriptomics Workflows.

The application of deep learning methods to spatial transcriptomics has shown promise in unraveling the complex relationships between gene expression patterns and tissue architecture as they pertain to various pathological conditions. Deep learning methods that can infer gene expression patterns directly from tissue histomorphology can expand the capability to discern spatial molecular markers within tissue slides. However, current methods utilizing these techniques are plagued by substantial variability in tissue preparation and characteristics, which can hinder the broader adoption of these tools. Furthermore, training deep learning models using spatial transcriptomics on small study cohorts remains a costly endeavor. Necessitating novel tissue preparation processes enhance assay reliability, resolution, and scalability. This study investigated the impact of an enhanced specimen processing workflow for facilitating a deep learning-based spatial transcriptomics assessment. The enhanced workflow leveraged the flexibility of the Visium CytAssist assay to permit automated H&E staining (e.g., Leica Bond) of tissue slides, whole-slide imaging at 40x-resolution, and multiplexing of tissue sections from multiple patients within individual capture areas for spatial transcriptomics profiling. Using a cohort of thirteen pT3 stage colorectal cancer (CRC) patients, we compared the efficacy of deep learning models trained on slide prepared using an enhanced workflow as compared to the traditional workflow which leverages manual tissue staining and standard imaging of tissue slides. Leveraging Inceptionv3 neural networks, we aimed to predict gene expression patterns across matched serial tissue sections, each stemming from a distinct workflow but aligned based on persistent histological structures. Findings indicate that the enhanced workflow considerably outperformed the traditional spatial transcriptomics workflow. Gene expression profiles predicted from enhanced tissue slides also yielded expression patterns more topologically consistent with the ground truth. This led to enhanced statistical precision in pinpointing biomarkers associated with distinct spatial structures. These insights can potentially elevate diagnostic and prognostic biomarker detection by broadening the range of spatial molecular markers linked to metastasis and recurrence. Future endeavors will further explore these findings to enrich our comprehension of various diseases and uncover molecular pathways with greater nuance. Combining deep learning with spatial transcriptomics provides a compelling avenue to enrich our understanding of tumor biology and improve clinical outcomes. For results of the highest fidelity, however, effective specimen processing is crucial, and fostering collaboration between histotechnicians, pathologists, and genomics specialists is essential to herald this new era in spatial transcriptomics-driven cancer research.