medRxiv - Pathology最新文献

{"title":"Oxidative stress and PARP activation in the lungs is an early event in COVID-19 pneumonia","authors":"Péter Juhász, Péter Bohus, Adrienn Sipos, Nicola Curtin, Gábor Méhes, Péter Bai","doi":"10.1101/2024.09.03.24312996","DOIUrl":"https://doi.org/10.1101/2024.09.03.24312996","url":null,"abstract":"Oxidative stress and poly(ADP-ribosyl)ation (PARylation) leads to tissue damage and inflammation in multiple lung diseases, likely in COVID-19. In a previous study we evidenced PARylation in multiple pulmonary cell types in patients who died of COVID-19, but not in patients who died of non-COVID-19 causes. We extended these observations in this retrospective immunohistochemical study by enlarging and stratifying the study population. We showed that pulmonary PARylation and oxidative stress peaked in the exudative and then decreased in the proliferative phase. Pulmonary oxidative stress and PARylation correlated with the serum markers of liver and kidney damage, oxygen transport, tissue hypoxia, lymphocytopenia, blood clotting and disseminated intravascular coagulation. Most correlations between PARylation and serum chemistry readouts were identified in the exudative phase. PARylation correlated with viral load and with the oxidative stress in the tissues, however, correlation between viral load and oxidative stress was marginal suggesting that oxidative stress and the presence of SARS-CoV-2 can independently induce PARylation. In males the time of hospitalization (time to death) was inversely correlated with pulmonary PARylation. Furthermore, males, died of COVID-19, were ∼15 years younger than females, however, there was no difference in pulmonary oxidative stress and PARylation between genders at death. Taken together, pulmonary PARylation and oxidative stress manifests early, in the exudative phase of COVID-19 and PARylation contributes to worse clinical outcome for males. These results suggest repurposing pharmacological PARP inhibitors for acute COVID-19 to counteract tissue damage.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"44 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142219070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PCSK9 inhibition in myeloid cells enhances cardioprotection beyond its LDL cholesterol-lowering effects","authors":"Shin Hye Moon, Imvastech Inc., Hyo Won Ki, Na Hyeon Yoon, Katherine I. Chung, Huiju Jo, Jing Jin, Sejin Jeon, Seung-Keun Sonn, Seungwoon Seo, Joowon Suh, Hyae Yon Kweon, Yun Seo Noh, Won Kee Yoon, Seung-Jun Lee, Chan Joo Lee, Nabil G Seidah, Sung Ho Park, Goo Taeg Oh","doi":"10.1101/2024.08.27.24312680","DOIUrl":"https://doi.org/10.1101/2024.08.27.24312680","url":null,"abstract":"BACKGROUND: Circulating levels of proprotein convertase subtilisin/kexin type 9 (PCSK9), which regulates plasma cholesterol content by degrading LDL receptor, are correlated with the risk of acute myocardial infarction (AMI). Recent studies suggested that PCSK9 improves cardiac function beyond its effects on LDL cholesterol levels after cardiac ischemic injury, but its precise mechanism remains unclear.\u0000METHODS: We examined the interrelationship and functional significance of PCSK9 and cardiac myeloid cells in ischemic hearts from AMI-induced Pcsk9-/- and Lyz2crePcsk9fl/fl mice, as well as in serum samples from coronary artery disease (CAD) patients treated with PCSK9 antibodies (Ab). Single-cell RNA sequencing (scRNA-seq) was conducted to identify heterogenous cardiac macrophage clusters and to investigate the impact of adaptive remodeling due to PCSK9 deficiency during AMI. Additionally, the regulatory effect of the myeloid-PCSK9/VEGF-C pathway was assessed in vitro as a potential therapeutic strategy.\u0000RESULTS: Our study demonstrated that PCSK9 deficiency induces diverse changes in myeloid cells and macrophages, potentially offering cardiac protection following AMI, irrespective of LDL cholesterol homeostasis. The scRNA-seq identified a subset of PCSK9-dependent cardiac macrophages (PDCMs) enriched in activator protein-1 (AP-1)?related pathways, functioning as reparative macrophages. These PDCMs were shown to enhance vascular endothelial growth factor C (VEGF-C) secretion and activate Akt signaling in cardiac endothelial cells, leading to improved cardiac remodeling. Notably, CAD patients treated with PCSK9 inhibitors exhibited increased numbers of myeloid cells with PDCM-like features, including elevated VEGF-C levels, consistent with our findings in mice.\u0000COUNCLUSIONS: Targeting PCSK9 in myeloid cells could offer cardioprotective effects by increasing AP-1 activity and VEGF-C expression of PDCMs, presenting a novel approach to preventing cardiac dysfunction in AMI. This strategy could expand the clinical use of existing PCSK9 inhibitors beyond just lowering LDL cholesterol.\u0000Key Words: coronary artery disease, lipid homeostasis, PCSK9 inhibitors, cardiac macrophages, activator protein-1, vascular endothelial growth factor C","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"14 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142219072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Camp Dream. Speak. Live.: A Virtual Adaptation","authors":"Courtney T Byrd, Geoffrey A Coalson, Danielle Werle, Robyn Croft, Katie L Winters, Megan M Young","doi":"10.1101/2024.08.22.24311294","DOIUrl":"https://doi.org/10.1101/2024.08.22.24311294","url":null,"abstract":"Purpose: The purpose of this study was to determine the efficacy of a virtual adaptation of the administration of Camp Dream. Speak. Live., an intensive, non-ableist manualized treatment program for children who stutter, with no indirect or direct fluency goals, in reducing the adverse impact of stuttering and increasing communication competence. Methods: Sixty-one children who stutter participated in Virtual Camp Dream. Speak. Live. Pre- and post-treatment measures were identical to previous in-person administrations: (1) self- and caregiver-report of cognitive and affective impact of stuttering (Communication Attitude Test for Children who Stutter [KiddyCAT/CAT], Overall Assessment of Speakers Experience of Stuttering [OASES], PROMIS Pediatric Peer Relationship, and PROMIS Parent Proxy Relationships), and (2) unfamiliar clinician ratings of communication competence of impromptu presentations. Results: Significant post-treatment gains were reported for the CAT, OASES, and PROMIS Peer Relationships Parent Proxy. Significant gains in post-treatment communication competence were observed. Pre-treatment stuttering frequency did not significantly predict changes in communication competence. Conclusion: Findings from Virtual Camp Dream. Speak. Live. demonstrate that the administration of the adapted telepractice format of this manualized program yields comparable findings as when administered in-person, suggesting promising implications for use in locations for which in-person provision and/or access is not feasible.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"20 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142219069","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Some statistical theory for interpreting reference distributions","authors":"Berk A Alpay, John M Higgins, Michael M Desai","doi":"10.1101/2024.07.23.24309680","DOIUrl":"https://doi.org/10.1101/2024.07.23.24309680","url":null,"abstract":"Reference distributions quantify the extremeness of clinical test results, typically relative to those of a healthy population. Intervals of these distributions are used in medical decision-making, but while there is much guidance for constructing them, the statistics of interpreting them for diagnosis have been less explored. Here we work directly in terms of the reference distribution, defining it as the likelihood in a posterior calculation of the probability of disease. We thereby identify assumptions of the conventional interpretation of reference distributions, criteria for combining tests, and considerations for personalizing interpretation of results from reference data. Theoretical reasoning supports that non-healthy variation be taken into account when possible, and that combining and personalizing tests call for careful statistical modeling.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141779734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical Relevance of Computationally Derived Tubular Features: Spatial Relationships and the Development of Tubulointerstitial Scarring in MCD/FSGS","authors":"Fan Fan, Qian Liu, Jarcy Zee, Takaya Ozeki, Dawit Demeke, Yingbao Yang, Alton B Farris, Bangchen Wang, Manav Shah, Jackson Jacobs, Laura Mariani, Kyle Lafata, Jeremy Rubin, Yijiang Chen, Lawrence Holzman, Jeffrey B Hodgin, Anant Madabhushi, Laura Barisoni, Andrew Janowczyk","doi":"10.1101/2024.07.19.24310619","DOIUrl":"https://doi.org/10.1101/2024.07.19.24310619","url":null,"abstract":"Background Visual scoring of tubular damage has limitations in capturing the full spectrum of structural changes and prognostic potential. We investigate if computationally quantified tubular features can enhance prognostication and reveal spatial relationships with interstitial fibrosis.\u0000Methods\u0000Deep-learning and image-processing-based segmentations were employed in N=254/266 PAS-WSIs from the NEPTUNE/CureGN datasets (135/153 focal segmental glomerulosclerosis and 119/113 minimal change disease) for: cortex, tubular lumen (TL), epithelium (TE), nuclei (TN), and basement membrane (TBM). N=104 pathomic features were extracted from these segmented tubular substructures and summarized at the patient level using summary statistics. The tubular features were quantified across the biopsy and in manually segmented regions of mature interstitial fibrosis and tubular atrophy (IFTA), pre-IFTA and non-IFTA in the NEPTUNE dataset. Minimum Redundancy Maximum Relevance was used in the NEPTUNE dataset to select features most associated with disease progression and proteinuria remission. Ridge-penalized Cox models evaluated their predictive discrimination compared to clinical/demographic data and visual-assessment. Models were evaluated in the CureGN dataset. Results\u0000N=9 features were predictive of disease progression and/or proteinuria remission. Models with tubular features had high prognostic accuracy in both NEPTUNE and CureGN datasets and increased prognostic accuracy for both outcomes (5.6%-7.7% and 1.6%-4.6% increase for disease progression and proteinuria remission, respectively) compared to conventional parameters alone in the NEPTUNE dataset. TBM thickness/area and TE simplification progressively increased from non- to pre- and mature IFTA.\u0000Conclusions\u0000Previously under-recognized, quantifiable, and clinically relevant tubular features in the kidney parenchyma can enhance understanding of mechanisms of disease progression and risk stratification.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"22 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141746233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Alzheimer's disease neuropathologic change mediates the relationship between ambient air pollution and dementia severity","authors":"Boram Kim, Kaitlin Blam, Holly Elser, Sharon X. Xie, Vivianna M. Van Deerlin, Trevor M. Penning, Daniel Weintraub, David J. Irwin, Lauren M. Massimo, Corey T. McMillan, Dawn Mechanic-Hamilton, David A. Wolk, Edward B. Lee","doi":"10.1101/2024.07.18.24310646","DOIUrl":"https://doi.org/10.1101/2024.07.18.24310646","url":null,"abstract":"Background: Exposure to fine particulate matter air pollution (PM2.5) increases risk for dementia. However, it is unknown whether this relationship is mediated by dementia-related neuropathologic change found at autopsy. We aimed to examine relationships between PM2.5 exposure, dementia severity, and dementia-associated neuropathologic change. Methods: This cross-sectional study used harmonized demographic, clinical, genetic, and neuropathological data from autopsy cases collected from 1998 to 2022 at the Center for Neurodegenerative Disease Research brain bank, University of Pennsylvania. Cases who had common neuropathologic forms of dementia and complete data on neuropathologic measures, APOE genotype, and residential address were included in this study cohort. Dementia severity was measured by Clinical Dementia Rating-Sum of Boxes (CDR-SB) scores. Ten dementia-associated neuropathologic measures representing Alzheimer's disease, Lewy body disease, limbic-predominant age related TDP-encephalopathy, and cerebrovascular disease were graded or staged according to the consensus criteria. One-year average PM2.5 exposure prior to death was estimated using a spatiotemporal prediction model based on residential addresses as the primary exposure measure. Linear, logistic and structural equation models were used to examine the relationships between PM2.5, CDR-SB and neuropathologic measures. Results: A total of 861 autopsy cases were included (mean age at death 76.6 years [SD 10.3]; 481 [56%] male). Each 1µg/m3 increase in one-year average PM2.5 concentration prior to death was associated with significantly greater cognitive and functional impairment (increase in CDR-SB score of 0.78; 95% confidence interval [CI], 0.52-1.05), faster cognitive and functional decline (change in CDR-SB scores of 0.13; 95% CI, 0.09-0.16), more severe Alzheimer's disease neuropathologic change (ADNC; odds ratio [OR] of 1.07; 95% CI, 1.01-1.13), and a higher prevalence of large infarcts (OR, 1.17; 95% CI, 1.05-1.30). The relationship between PM2.5 exposure and CDR-SB was mediated by ADNC (change in CDR-SB score due to ADNC level of 0.36; 95% CI, 0.13-0.65). Conclusions: PM2.5 exposure may increase dementia risk by increasing ADNC. Measures that improve air quality may represent a population-level intervention for the prevention of dementia.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"43 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141744837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Large Language Models in Pathology: A Comparative Study on Multiple Choice Question Performance with Pathology Trainees","authors":"Wei Du, Jaryse Harris, Alessandro Brunetti, Olivia Leung, Xingchen Li, Selemon Walle, Qing Yu, Xiao Zhou, Fang Bian, Kajanna Mckenzie, Xueting Jin, Manita Kanathanavanich, Farah El-Sharkawy, Shunsuke Koga","doi":"10.1101/2024.07.10.24310093","DOIUrl":"https://doi.org/10.1101/2024.07.10.24310093","url":null,"abstract":"Aims: Large language models (LLMs), such as ChatGPT and Bard, have shown potential in various medical applications. This study aims to evaluate the performance of LLMs, specifically ChatGPT and Bard, in pathology by comparing their performance with that of pathology residents and fellows, and to assess the consistency of their responses.\u0000Methods: We selected 150 multiple-choice questions covering 15 subspecialties, excluding those with images. Both ChatGPT and Bard were tested on these questions three times, and their responses were compared with those of 14 pathology trainees from two hospitals. Questions were categorized into easy, intermediate, and difficult based on trainee performance. Consistency and variability in LLM responses were analyzed across three evaluation sessions.\u0000Results: ChatGPT significantly outperformed Bard and trainees, achieving an average total score of 82.2% compared to Bard's 49.5% and trainees' 50.7%. ChatGPT's performance was notably stronger in difficult questions (61.8%-70.6%) compared to Bard (29.4%-32.4%) and trainees (5.9%-44.1%). For easy questions, ChatGPT (88.9%-94.4%) and trainees (75.0%-100.0%) showed similar high scores. Consistency analysis revealed that ChatGPT showed a high consistency rate of 85%-80% across three tests, whereas Bard exhibited greater variability with consistency rates of 61%-54%.\u0000Conclusion: ChatGPT consistently outperformed Bard and trainees, especially on difficult questions. While LLMs show significant potential in pathology education and practice, ongoing development and human oversight are essential for reliable clinical application.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"28 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141613244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the power of MRI and clinical measures in predicting Alzheimers disease neuropathology","authors":"Farooq Kamal, Cassandra Morrison, Michael D. Oliver, Mahsa Dadar","doi":"10.1101/2024.07.09.24310163","DOIUrl":"https://doi.org/10.1101/2024.07.09.24310163","url":null,"abstract":"Background: The ability to predict Alzheimers disease (AD) before diagnosis is a topic of intense research. Early diagnosis would aid in improving treatment and intervention options, however, there are no current methods that can accurately predict AD years in advance. This study examines a novel machine learning approach that integrates the combined effects of vascular (white matter hyperintensities, WMHs), and structural brain changes (gray matter, GM) with clinical factors (cognitive status) to predict post-mortem neuropathological outcomes. Methods: Healthy older adults, participants with mild cognitive impairment, and AD from the Alzheimer's Disease Neuroimaging Initiative dataset with both post-mortem neuropathology data and antemortem MRI and clinical data were included. Longitudinal data were analyzed across three intervals before death (post-mortem data): 0-4 years, 4-8 years, and 8-14 years. Additionally, cross-sectional data at the last visit or interval (within four years, 0-4 years) before death were also examined. Machine learning models including gradient boosting, bagging, support vector regression, and linear regression were implemented. These models were applied towards feature selection of the top seven MRI, clinical, and demographic data to identify the best performing set of variables that could predict postmortem neuropathology outcomes (i.e., neurofibrillary tangles, neuritic plaques, diffuse plaques, senile/amyloid plaques, and amyloid angiopathy). Results: A total of 94 participants (55-90 years of age) were included in the study. At last visit, the best-performing model included total and temporal lobe WMHs and achieved r=0.87(RMSE=0.62) during cross-validation for neuritic plaques. For longitudinal assessments across different intervals, the best-performing model included regional GM (i.e., hippocampus, amygdala, caudate) and frontal lobe WMH and achieved r=0.93(RMSE=0.59) during cross-validation for neurofibrillary tangles. For MRI and clinical predictors and clinical-only predictors, t-tests demonstrated significant differences at all intervals before death (t[-13.60-7.90], p-values<0.001). Overall, post-mortem neuropathology outcome were predicted up to 14 years before death with high accuracies (~90%).\u0000Conclusions: Prediction accuracy was higher for post-mortem neuropathology outcomes that included MRI (WMHs, GM) and clinical features compared to clinical-only features. These findings highlight that MRI features are critical to successfully predict AD-related pathology years in advance which will improve participant selection for clinical trials, treatments, and intervention options.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"26 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141584879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust, credible, and interpretable AI-based histopathological prostate cancer grading","authors":"Fabian Westhaeusser, Patrick Fuhlert, Esther Dietrich, Maximilian Lennartz, Robin Khatri, Nico Kaiser, Pontus Roebeck, Roman Buelow, Saskia von Stillfried, Anja Witte, Sam Ladjevardi, Anders Drotte, Peter Severgardh, Jan Baumbach, Victor G Puelles, Michael Haeggman, Michael Brehler, Peter Boor, Peter Walhagen, Anca Dragomir, Christer Busch, Markus Graefen, Ewert Bengtsson, Guido Sauter, Marina Zimmermann, Stefan Bonn","doi":"10.1101/2024.07.09.24310082","DOIUrl":"https://doi.org/10.1101/2024.07.09.24310082","url":null,"abstract":"Background: Prostate cancer (PCa) is among the most common cancers in men and its diagnosis requires the histopathological evaluation of biopsies by human experts. While several recent artificial intelligence-based (AI) approaches have reached human expert-level PCa grading, they often display significantly reduced performance on external datasets. This reduced performance can be caused by variations in sample preparation, for instance the staining protocol, section thickness, or scanner used. Another limiting factor of contemporary AI-based PCa grading is the prediction of ISUP grades, which leads to the perpetuation of human annotation errors. Methods: We developed the prostate cancer aggressiveness index (PCAI), an AI-based PCa detection and grading framework that is trained on objective patient outcome, rather than subjective ISUP grades. We designed PCAI as a clinical application, containing algorithmic modules that offer robustness to data variation, medical interpretability, and a measure of prediction confidence. To train and evaluate PCAI, we generated a multicentric, retrospective, observational trial consisting of six cohorts with 25,591 patients, 83,864 images, and 5 years of median follow-up from 5 different centers and 3 countries. This includes a high-variance dataset of 8,157 patients and 28,236 images with variations in sample thickness, staining protocol, and scanner, allowing for the systematic evaluation and optimization of model robustness to data variation. The performance of PCAI was assessed on three external test cohorts from two countries, comprising 2,255 patients and 9,437 images. Findings: Using our high-variance datasets, we show how differences in sample processing, particularly slide thickness and staining time, significantly reduce the performance of AI-based PCa grading by up to 6.2 percentage points in the concordance index (C-index). We show how a select set of algorithmic improvements, including domain adversarial training, conferred robustness to data variation, interpretability, and a measure of credibility to PCAI. These changes lead to significant prediction improvement across two biopsy cohorts and one TMA cohort, systematically exceeding expert ISUP grading in C-index and AUROC by up to 22 percentage points. Interpretation: Data variation poses serious risks for AI-based histopathological PCa grading, even when models are trained on large datasets. Algorithmic improvements for model robustness, interpretability, credibility, and training on high-variance data as well as outcome-based severity prediction gives rise to robust models with above ISUP-level PCa grading performance.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"87 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141571348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accurate diagnosis achieved via super-resolution whole slide images by pathologists and artificial intelligence","authors":"kuansong wang, Ruijie LIU, Yushi chen, ying wang, yanning qiu, yanhua gao, maoxu zhou, bingqian bai, mingxiang zhang, kai sun, Hong-Wen Deng, hongmei xiao, Gang Yu","doi":"10.1101/2024.07.05.24310022","DOIUrl":"https://doi.org/10.1101/2024.07.05.24310022","url":null,"abstract":"Background: Digital pathology significantly improves diagnostic efficiency and accuracy; however, pathological tissue sections are scanned at high resolutions (HR), magnified by 40 times (40X) incurring high data volume, leading to storage bottlenecks for processing large numbers of whole slide images (WSIs) for later diagnosis in clinic and hospitals.\u0000Method: We propose to scan at a magnification of 5 times (5X). We developed a novel multi-scale deep learning super-resolution (SR) model that can be used to accurately computes 40X SR WSIs from the 5X WSIs. Results: The required storage size for the resultant data volume of 5X WSIs is only one sixty-fourth (less than 2%) of that of 40X WSIs. For comparison, three pathologists used 40X scanned HR and 40X computed SR WSIs from the same 480 histology glass slides spanning 47 diseases (such tumors, inflammation, hyperplasia, abscess, tumor-like lesions) across 12 organ systems. The results are nearly perfectly consistent with each other, with Kappa values (HR and SR WSIs) of 0.988±0.018, 0.924±0.059, and 0.966±0.037, respectively, for the three pathologists. There were no significant differences in diagnoses of three pathologists between the HR and corresponding SR WSIs, with Area under the Curve (AUC): 0.920±0.164 vs. 0.921±0.158 (p-value=0.653), 0.931±0.128 vs. 0.943±0.121 (p-value=0.736), and 0.946±0.088 vs. 0.941±0.098 (p-value=0.198). A previously developed highly accurate colorectal cancer artificial intelligence system (AI) diagnosed 1,821 HR and 1,821 SR WSIs, with AUC values of 0.984±0.016 vs. 0.984±0.013 (p-value=0.810), again with nearly perfect matching results.\u0000Conclusions: The pixel numbers of 5X WSIs is only less than 2% of that of 40X WSIs. The 40X computed SR WSIs can achieve accurate diagnosis comparable to 40X scanned HR WSIs, both by pathologists and AI. This study provides a promising solution to overcome a common storage bottleneck in digital pathology.","PeriodicalId":501528,"journal":{"name":"medRxiv - Pathology","volume":"28 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141571349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}