Gastric Cancer Detection with Ensemble Learning on Digital Pathology: Use Case of Gastric Cancer on GasHisSDB Dataset.

Gastric cancer has become a serious worldwide health concern, emphasizing the crucial importance of early diagnosis measures to improve patient outcomes. While traditional histological image analysis is regarded as the clinical gold standard, it is labour intensive and manual. In recognition of this problem, there has been a rise in interest in the use of computer-aided diagnostic tools to help pathologists with their diagnostic efforts. In particular, deep learning (DL) has emerged as a promising solution in this sector. However, current DL models are still restricted in their ability to extract extensive visual characteristics for correct categorization. To address this limitation, this study proposes the use of ensemble models, which incorporate the capabilities of several deep-learning architectures and use aggregate knowledge of many models to improve classification performance, allowing for more accurate and efficient gastric cancer detection. To determine how well these proposed models performed, this study compared them with other works, all of which were based on the Gastric Histopathology Sub-Size Images Database, a publicly available dataset for gastric cancer. This research demonstrates that the ensemble models achieved a high detection accuracy across all sub-databases, with an average accuracy exceeding 99%. Specifically, ResNet50, VGGNet, and ResNet34 performed better than EfficientNet and VitNet. For the 80 × 80-pixel sub-database, ResNet34 exhibited an accuracy of approximately 93%, VGGNet achieved 94%, and the ensemble model excelled with 99%. In the 120 × 120-pixel sub-database, the ensemble model showed 99% accuracy, VGGNet 97%, and ResNet50 approximately 97%. For the 160 × 160-pixel sub-database, the ensemble model again achieved 99% accuracy, VGGNet 98%, ResNet50 98%, and EfficientNet 92%, highlighting the ensemble model's superior performance across all resolutions. Overall, the ensemble model consistently provided an accuracy of 99% across the three sub-pixel categories. These findings show that ensemble models may successfully detect critical characteristics from smaller patches and achieve high performance. The findings will help pathologists diagnose gastric cancer using histopathological images, leading to earlier identification and higher patient survival rates.

Multi-Scale Digital Pathology Patch-Level Prostate Cancer Grading Using Deep Learning: Use Case Evaluation of DiagSet Dataset.

Prostate cancer remains a prevalent health concern, emphasizing the critical need for early diagnosis and precise treatment strategies to mitigate mortality rates. The accurate prediction of cancer grade is paramount for timely interventions. This paper introduces an approach to prostate cancer grading, framing it as a classification problem. Leveraging ResNet models on multi-scale patch-level digital pathology and the Diagset dataset, the proposed method demonstrates notable success, achieving an accuracy of 0.999 in identifying clinically significant prostate cancer. The study contributes to the evolving landscape of cancer diagnostics, offering a promising avenue for improved grading accuracy and, consequently, more effective treatment planning. By integrating innovative deep learning techniques with comprehensive datasets, our approach represents a step forward in the pursuit of personalized and targeted cancer care.

Ensemble Deep Learning-Based Image Classification for Breast Cancer Subtype and Invasiveness Diagnosis from Whole Slide Image Histopathology.

Cancer diagnosis and classification are pivotal for effective patient management and treatment planning. In this study, a comprehensive approach is presented utilizing ensemble deep learning techniques to analyze breast cancer histopathology images. Our datasets were based on two widely employed datasets from different centers for two different tasks: BACH and BreakHis. Within the BACH dataset, a proposed ensemble strategy was employed, incorporating VGG16 and ResNet50 architectures to achieve precise classification of breast cancer histopathology images. Introducing a novel image patching technique to preprocess a high-resolution image facilitated a focused analysis of localized regions of interest. The annotated BACH dataset encompassed 400 WSIs across four distinct classes: Normal, Benign, In Situ Carcinoma, and Invasive Carcinoma. In addition, the proposed ensemble was used on the BreakHis dataset, utilizing VGG16, ResNet34, and ResNet50 models to classify microscopic images into eight distinct categories (four benign and four malignant). For both datasets, a five-fold cross-validation approach was employed for rigorous training and testing. Preliminary experimental results indicated a patch classification accuracy of 95.31% (for the BACH dataset) and WSI image classification accuracy of 98.43% (BreakHis). This research significantly contributes to ongoing endeavors in harnessing artificial intelligence to advance breast cancer diagnosis, potentially fostering improved patient outcomes and alleviating healthcare burdens.

Distance to Treatment With Radical Cystectomy in a Rural State: Long Car Rides, Equivalent Outcomes.

INTRODUCTION: Radical cystectomy is a complex surgery with better outcomes reported when performed at high-volume centers. This may lead to patients traveling farther for care. We examined the impact of travel distance on clinical outcomes. METHODS: A total of 220 patients undergoing radical cystectomy from 2015-2021 were retrospectively reviewed. Distance traveled to the treatment center by patient zip codes was classified as <12.5 miles, 12.5-49.9 miles, and ≥50 miles. Multivariable logistic regression was used to assess complications, readmissions, 90-day mortality, and length of stay by distance traveled. Time to treatment based on distance traveled was compared. RESULTS: A total of 220 patients underwent radical cystectomy with complete 90-day follow-up. Of the patients 38.6% (85/220) were readmitted; 62.5% (53/85) presented to the treatment center or were transferred. All patients readmitted to an outside hospital traveled ≥12.5 miles (P < .001). Patients with high-grade complications were likely to be transferred to the treatment center with only 23.7% (9/38) definitively managed by outside hospital. Patients traveling >12.5 miles with low-grade complications were more likely to be managed at an outside hospital (57.5%, P = .01). There was no difference in time to initiation of neoadjuvant chemotherapy (P = .99) or time to radical cystectomy following neoadjuvant chemotherapy (P = .23) by distance traveled. For 49 muscle-invasive bladder cancer patients proceeding directly to surgery without neoadjuvant chemotherapy, time from diagnosis to radical cystectomy was increased if traveling >12.5 miles (P = .04). CONCLUSIONS: Increased travel distance did not impact early postoperative outcomes. Distance traveled may impact access to care, such as time to surgery or location of readmission to the treatment center postoperatively.

Prostate Specific Antigen Testing Behaviors for Prostate Cancer Screening Among U.S. Immigrants: A Cross-sectional Analysis Using the National Health Interview Survey.

PURPOSE: Immigrants constitute 14% of the U.S. population, and this group is especially vulnerable to poor health care access. Prior research demonstrates U.S. immigrants have low rates of guideline-concordant breast and colorectal screening, but prostate cancer screening has not previously been evaluated. We sought to characterize screening behaviors among U.S. immigrants and to consider possible mechanisms to enhance PSA-based screening for this population. MATERIALS AND METHODS: Data were obtained from the 2010, 2013, 2015, and 2018 National Health Interview Survey reports, which were the recent survey years that included questions about PSA testing. Complex samples logistic regression was performed to assess the relationship between immigrant-specific characteristics including region of birth, citizenship status, length of residence within the U.S., English language proficiency, and history of PSA testing. RESULTS: There were 22,997 survey respondents; 3,257 were foreign-born and 19,740 were U.S.-born. Rates of PSA testing were much lower among the foreign-born population compared to the U.S.-born population (43% vs 60%). Citizenship status, length of residence in the U.S. for more than 15 years, and English proficiency were directly linked to increased rates of PSA testing. There was significant variability in PSA testing among immigrant subgroups and Asian immigrants had the lowest rate of PSA testing. Annual physician visits and English language proficiency were associated with increased PSA testing among the U.S. immigrant population. CONCLUSIONS: Immigrants have relatively low rates of PSA testing. Improving health care utilization and language services may help to narrow the gap in guideline-concordant prostate cancer screening between immigrants and nonimmigrants.

PRState: Incorporating genetic ancestry in prostate cancer risk scores for men of African ancestry.

BACKGROUND: Prostate cancer (PrCa) is one of the most genetically driven solid cancers with heritability estimates as high as 57%. Men of African ancestry are at an increased risk of PrCa; however, current polygenic risk score (PRS) models are based on European ancestry groups and may not be broadly applicable. The objective of this study was to construct an African ancestry-specific PrCa PRS (PRState) and evaluate its performance. METHODS: African ancestry group of 4,533 individuals in ELLIPSE consortium was used for discovery of African ancestry-specific PrCa SNPs. PRState was constructed as weighted sum of genotypes and effect sizes from genome-wide association study (GWAS) of PrCa in African ancestry group. Performance was evaluated using ROC-AUC analysis. RESULTS: We identified African ancestry-specific PrCa risk loci on chromosomes 3, 8, and 11 and constructed a polygenic risk score (PRS) from 10 African ancestry-specific PrCa risk SNPs, achieving an AUC of 0.61 [0.60-0.63] and 0.65 [0.64-0.67], when combined with age and family history. Performance dropped significantly when using ancestry-mismatched PRS models but remained comparable when using trans-ancestry models. Importantly, we validated the PRState score in the Million Veteran Program (MVP), demonstrating improved prediction of PrCa and metastatic PrCa in individuals of African ancestry. CONCLUSIONS: African ancestry-specific PRState improves PrCa prediction in African ancestry groups in ELLIPSE consortium and MVP. This study underscores the need for inclusion of individuals of African ancestry in gene variant discovery to optimize PRSs and identifies African ancestry-specific variants for use in future studies.

On-Clamp vs. Off-Clamp Robot-Assisted Partial Nephrectomy for cT2 Renal Tumors: Retrospective Propensity-Score-Matched Multicenter Outcome Analysis.

We compared perioperative outcomes after on-clamp versus off-clamp robot-assisted partial nephrectomy (RAPN) for >7 cm renal masses. A multicenter dataset was queried for patients who had undergone RAPN for a cT2cN0cM0 kidney tumor from July 2007 to February 2022. The Trifecta achievement (negative surgical margins, no severe complications, and ≤ 30% postoperative estimated glomerular filtration rate (eGFR) reduction) was considered a surrogate of surgical quality. Overall, 316 cases were included in the analysis, and 58% achieved the Trifecta. A propensity-score-matched analysis generated two cohorts of 89 patients homogeneous for age, ASA score, preoperative eGFR, and RENAL score (all p > 0.21). Compared to the on-clamp approach, OT was significantly shorter in the off-clamp group (80 vs. 190 min; p < 0.001), the incidence of sRFD was lower (22% vs. 40%; p = 0.01), and the Trifecta rate higher (66% vs. 46%; p = 0.01). In a crude analysis, >20 min of hilar clamping was associated with a significantly higher risk of sRFD (OR: 2.30; 95%CI: 1.13−4.64; p = 0.02) and with reduced probabilities of achieving the Trifecta (OR: 0.46; 95%CI: 0.27−0.79; p = 0.004). Purely off-clamp RAPN seems to be a safe and viable option to treat cT2 renal masses and may outperform the on-clamp approach regarding perioperative surgical outcomes.

Clarifying the Trade-Offs of Risk-Stratified Screening for Prostate Cancer: A Cost-Effectiveness Study.

Cancer risk prediction is necessary for precision early detection, which matches screening intensity to risk. However, practical steps for translating risk predictions to risk-stratified screening policies are not well established. We used a validated population prostate-cancer model to simulate the outcomes of strategies that increase intensity for men at high risk and reduce intensity for men at low risk. We defined risk by the Prompt Prostate Genetic Score (PGS) (Stratify Genomics, San Diego, California), a germline genetic test. We first recalibrated the model to reflect the disease incidence observed within risk strata using data from a large prevention trial where some participants were tested with Prompt PGS. We then simulated risk-stratified strategies in a population with the same risk distribution as the trial and evaluated the cost-effectiveness of risk-stratified screening versus universal (risk-agnostic) screening. Prompt PGS risk-adapted screening was more cost-effective when universal screening was conservative. Risk-stratified strategies improved outcomes at a cost of less than $100,000 per quality-adjusted life year compared with biennial screening starting at age 55 years, but risk stratification was not cost-effective compared with biennial screening starting at age 45. Heterogeneity of risk and fraction of the population within each stratum were also important determinants of cost-effectiveness.

National Recommendations against Prostate Specific Antigen Screening versus Statewide Medicaid Expansion Initiatives: A Battle of the Titans.

PURPOSE: Medicaid expansion under the Patient Protection and Affordable Care Act occurred almost concurrently with 2012 U.S. Preventive Services Task Force recommendations against prostate specific antigen screening. Here the relative influence on prostate specific antigen screening rates by 2 concurrent and opposing system-level policy initiatives is investigated: improved access to care and change in clinical practice guidelines. MATERIALS AND METHODS: Behavioral Risk Factor Surveillance System data from years 2012 to 2018 were analyzed for trends in self-reported prostate specific antigen screening and insurance coverage. Subanalyses included state Medicaid expansion status and respondent federal poverty level. Multivariable logistic regression was performed to evaluate factors associated with prostate specific antigen screening. RESULTS: From 2012 to 2018 prostate specific antigen screening predominantly declined with a notable exception of an increase of 7.3% for men at <138% federal poverty level between 2011 and 2013 in early expansion states. Initial increases did not continue, and screening trends mirrored those of nonexpansion states by 2018. Notably, 2014 planned expansions states did not follow this trend with minimal change between 2015 and 2017 compared to declines in early expansion states and nonexpansion states (-0.4% vs -6.7% and -8.6%, respectively). CONCLUSIONS: Medicaid expansion was associated with increased rates of insured men at <138% federal poverty level from 2012 to 2018 in early expansion states. In this group, initial increases in prostate specific antigen screening were not durable and followed the trend of reduced screening seen across the United States. In planned expansions states the global drop in prostate specific antigen screening from 2016 to 2018 was offset in men at <138% federal poverty level by expanding access to care. Nonexpansion states showed a steady decline in prostate specific antigen screening rates. This suggests that policy such as U.S. Preventive Services Task Force recommendations against screening competes with and often outmatches access to care.