Automatic removal of large blood vasculature for objective assessment of brain tumors using quantitative dynamic contrast-enhanced magnetic resonance imaging.

The presence of a normal large blood vessel (LBV) in a tumor region can impact the evaluation of quantitative dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parameters and tumor classification. Hence, there is a need for automatic removal of LBVs from brain tissues including intratumoral regions for achieving an objective assessment of tumors. This retrospective study included 103 histopathologically confirmed brain tumor patients who underwent MRI, including DCE-MRI data acquisition. Quantitative DCE-MRI analysis was performed for computing various parameters such as wash-out slope (Slope-2), relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), blood plasma volume fraction (Vp), and volume transfer constant (Ktrans). An approach based on data-clustering algorithm, morphological operations, and quantitative DCE-MRI maps was proposed for the segmentation of normal LBVs in brain tissues, including the tumor region. Here, three widely used data-clustering algorithms were evaluated on two types of quantitative maps: (a) Slope-2, and (b) a new proposed combination of rCBV and Slope-2 maps. Fluid-attenuated inversion recovery-MRI hyperintense lesions were also automatically segmented using deep learning-based architecture. The accuracy of LBV segmentation was qualitatively assessed blindly by two experienced observers, and Likert scoring was also obtained from each individual and compared using Cohen's Kappa test, and multiple statistical features from quantitative DCE-MRI parameters were obtained in the segmented tumor. t-test and receiver operating characteristic (ROC) curve analysis were performed for comparing the effect of removal of LBVs on parameters as well as on tumor grading. k-means clustering exhibited better accuracy and computational efficiency. Tumors, in particular high-grade gliomas (HGGs), showed a high contrast compared with normal tissues (relative % difference = 18.5%) on quantitative maps after the removal of LBVs. Statistical features (95th percentile values) of all parameters in the tumor region showed a statistically significant difference (p < 0.05) between with and without LBV maps. Similar results were obtained for the ROC curve analysis for differentiation between low-grade gliomas and HGGs. Moreover, after the removal of LBVs, the rCBV, rCBF, and Vp maps show better visualization of tumor regions.

Predictive factors for recurrence and outcomes in T1a renal cell carcinoma: Analysis of the INMARC (International Marker Consortium for Renal Cancer) database.

OBJECTIVE: Stage migration in renal cell carcinoma (RCC) has led to an increasing proportion of diagnosed small renal masses. Emerging knowledge regarding heterogeneity of RCC histologies and consequent impact on prognosis led us to further explore outcomes and predictive factors in surgically-treated T1a RCC. METHODS: The INMARC database was queried for T1aN0M0 RCC. Patients were stratified into groups based on recurrence. Primary outcome was overall survival (OS). Multivariable analyses (MVA) were performed for factors associated with recurrence, cancer-specific (CSM), and all-cause mortality (ACM). Kaplan-Meier analyses (KMA) assessed survival by histology and grade. Subset analysis for time to recurrence was conducted for grade and histologic groups and compared with recent AUA follow-up guidelines [low-risk (AUA-LR), intermediate-risk (AUA-IR), high-risk (AUA-HR), and very-high risk (AUA-VHR) groups]. RESULTS: We analyzed 1,878 patients (median follow-up 35.2 months); 101 (5.4%) developed recurrence. MVA for recurrence demonstrated increasing age (P = 0.026), male sex (P = 0.043), diabetes (P = 0.007), high/unclassified grade (P < 0.001-0.007), and variant histology (P = 0.017) as independent risk factors for increased risk, while papillary (P = 0.016) and chromophobe (P = 0.049) were associated with decreased risk. MVA identified high/unclassified grade (P = 0.003-0.004) and pT3a upstaging (P = 0.043) as predictive factors for worsened risk of CSM while papillary (P = 0.034) was associated with improved risk. MVA for ACM demonstrated increasing age (P < 0.001), non-white (P < 0.001), high-grade (P = 0.022), variant histology (P = 0.049), recurrence (P = 0.004), and eGFR<45 at last follow-up (P < 0.001) to be independent risk factors. KMA comparing clear cell, chromophobe, papillary, and variant RCC revealed significant differences for 5-year CSS (P = 0.018) and RFS (P < 0.001), but not OS (P = 0.34). Median time to recurrence was 23.8 months for low-grade (AUA-LR), 17.3 months for high-grade (AUA-IR), 18 months for pT3a upstaging (AUA-HR), and 12 months for variant histology (AUA-VHR; P < 0.001). CONCLUSION: We noted differential outcomes in T1a RCC based on histology and grade for recurrence and CSM, while renal functional decline in addition to pathological factors and recurrence were predictive for ACM. Our findings support recently promulgated AUA follow-up guidelines for low-grade and variant histology pT1a RCC, but call for consolidation of follow-up protocols for high-grade pT1a and pT3a upstaged patients, with intensification of frequency of imaging follow-up in pT1a high-grade RCC.

Diagnostic accuracy of upper tract urothelial carcinoma using biopsy, urinary cytology, and nephroureterectomy specimens: A tertiary cancer center experience.

OBJECTIVES: We studied the diagnostic accuracy and discordance of upper tract urothelial carcinoma (UTUC) by comparing biopsy and urinary cytology with matched nephroureterectomy specimens. METHODS: Sixty-nine patients with UTUC without neoadjuvant treatment were retrospectively identified who had matched biopsy and nephroureterectomy specimens. Twenty patients had concurrent upper tract cytology. H&E and cytology slides were re-reviewed. Statistical analysis was performed. RESULTS: Patients included 48 men and 21 women with a mean age of 69 years. A concordant grade between biopsy and surgical specimen was present in 49 (71%) patients. The mean size of biopsy specimens in the discordant group was significantly smaller than that in the concordant group. Invasion was evaluated in 48 biopsy cases that had adequate subepithelial tissue, and 33 of them were diagnosed with concordant invasion status. Mean tumor size in both tumor grade and invasion discordant groups was significantly larger than that in the concordant group. High-grade urothelial carcinoma was detected in 84% of cases using urinary cytology. CONCLUSIONS: Our study demonstrates the diagnostic challenges of UTUC on small biopsy specimens. Biopsy specimen size and tumor size are significantly associated with the diagnostic discordance. Upper tract cytology showed high diagnostic accuracy and should be complementary to the biopsy.

Analyzing Tumor Budding Scores in Invasive Breast Carcinoma: A Tertiary Care Center Study in South India.

Introduction and Aim: Tumor budding is a distinctive phenomenon which involves the presence of small clusters or individual cancer cells at the invasive front of tumors. Tumor budding has garnered attention due to its potential implications for prognosis, treatment strategies, and our understanding of cancer progression. Our aim is to study the distribution of tumor buds and its scoring in patients with infiltrating breast carcinoma and to associate with other histopathological parameters like the size of the tumor, its grade, lymphovascular invasion, and lymph node metastasis. Materials and Methods: This was a study analyzing the data of 70 resected specimens of primary breast carcinomas and providing a descriptive overview. Tumor budding was recognized, counted, and graded in hematoxylin and eosin slides. The cases were classified as low (0-4), intermediate (5-9), and high (≥10 buds) based on the count of tumor buds. Tumor budding has significant correlation with tumor grade and tumor size. Results: Of the 70 cases, 60 cases (85.71%) were diagnosed as invasive ductal carcinoma NOS. The majority [38 (54.28%)] of the cases showed an intermediate tumor budding score of 5-9/10 HPF. Conclusion: Evaluation of tumor budding allows pathologists and oncologists to gather valuable information about the tumor's biological aggressiveness and potential for metastasis. It also helps in better risk stratification of patients, enabling a more personalized and tailored approach to treatment planning. In conclusion, assessing tumor budding in breast carcinoma holds significant clinical importance in the management and prognosis of this disease.

Novel Insights Into the International Association for the Study of Lung Cancer Grading System for Lung Adenocarcinoma.

The new grading system for lung adenocarcinoma proposed by the International Association for the Study of Lung Cancer (IASLC) defines prognostic subgroups on the basis of histologic patterns observed on surgical specimens. This study sought to provide novel insights into the IASLC grading system, with particular focus on recurrence-specific survival (RSS) and lung cancer-specific survival among patients with stage I adenocarcinoma. Under the IASLC grading system, tumors were classified as grade 1 (lepidic predominant with <20% high-grade patterns [micropapillary, solid, and complex glandular]), grade 2 (acinar or papillary predominant with <20% high-grade patterns), or grade 3 (≥20% high-grade patterns). Kaplan-Meier survival estimates, pathologic features, and genomic profiles were investigated for patients whose disease was reclassified into a higher grade under the IASLC grading system on the basis of the hypothesis that they would strongly resemble patients with predominant high-grade tumors. Overall, 423 (29%) of 1443 patients with grade 1 or 2 tumors classified based on the predominant pattern-based grading system had their tumors upgraded to grade 3 based on the IASLC grading system. The RSS curves for patients with upgraded tumors were significantly different from those for patients with grade 1 or 2 tumors (log-rank P < .001) but not from those for patients with predominant high-grade patterns (P = .3). Patients with upgraded tumors had a similar incidence of visceral pleural invasion and spread of tumor through air spaces as patients with predominant high-grade patterns. In multivariable models, the IASLC grading system remained significantly associated with RSS and lung cancer-specific survival after adjustment for aggressive pathologic features such as visceral pleural invasion and spread of tumor through air spaces. The IASLC grading system outperforms the predominant pattern-based grading system and appropriately reclassifies tumors into higher grades with worse prognosis, even after other pathologic features of aggressiveness are considered.

Invasion and Grading of Pulmonary Non-Mucinous Adenocarcinoma.

Lung adenocarcinoma staging and grading were recently updated to reflect the link between histologic growth patterns and outcomes. The lepidic growth pattern is regarded as "in-situ," whereas all other patterns are regarded as invasive, though with stratification. Solid, micropapillary, and complex glandular patterns are associated with worse prognosis than papillary and acinar patterns. These recent changes have improved prognostic stratification. However, multiple pitfalls exist in measuring invasive size and in classifying lung adenocarcinoma growth patterns. Awareness of these limitations and recommended practices will help the pathology community achieve consistent prognostic performance and potentially contribute to improved patient management.

Dual path parallel hierarchical diagnosis model for intracranial tumors based on multi-feature entropy weight.

The grading diagnosis of intracranial tumors is a key step in formulating clinical treatment plans and surgical guidelines. To effectively grade the diagnosis of intracranial tumors, this paper proposes a dual path parallel hierarchical model that can automatically grade the diagnosis of intracranial tumors with high accuracy. In this model, prior features of solid tumor mass and intratumoral necrosis are extracted. Then the optimal division of the data set is achieved through multi-feature entropy weight. The multi-modal input is realized by the dual path structure. Multiple features are superimposed and fused to achieve the image grading. The model has been tested on the actual clinical medical images provided by the Second Affiliated Hospital of Dalian Medical University. The experiment shows that the proposed model has good generalization ability, with an accuracy of 0.990. The proposed model can be applied to clinical diagnosis and has practical application prospects.

Deep learning-assisted survival prognosis in renal cancer: A CT scan-based personalized approach.

This paper presents a deep learning (DL) approach for predicting survival probabilities of renal cancer patients based solely on preoperative CT imaging. The proposed approach consists of two networks: a classifier- and a survival- network. The classifier attempts to extract features from 3D CT scans to predict the ISUP grade of Renal cell carcinoma (RCC) tumors, as defined by the International Society of Urological Pathology (ISUP). Our classifier is a 3D convolutional neural network to avoid losing crucial information on the interconnection of slides in 3D images. We employ multiple procedures, including image augmentation, preprocessing, and concatenation, to improve the performance of the classifier. Given the strong correlation between ISUP grading and renal cancer prognosis in the clinical context, we use the ISUP grading features extracted by the classifier as the input to the survival network. By leveraging this clinical association and the classifier network, we are able to model our survival analysis using a simple DL-based network. We adopt a discrete LogisticHazard-based loss to extract intrinsic survival characteristics of RCC tumors from CT images. This allows us to build a completely parametric survival model that varies with patients' tumor characteristics and predicts non-proportional survival probability curves for different patients. Our results demonstrated that the proposed method could predict the future course of renal cancer with reasonable accuracy from the CT scans. The proposed method obtained an average concordance index of 0.72, an integrated Brier score of 0.15, and an area under the curve value of 0.71 on the test cohorts.

A hybrid deep CNN model for brain tumor image multi-classification.

The current approach to diagnosing and classifying brain tumors relies on the histological evaluation of biopsy samples, which is invasive, time-consuming, and susceptible to manual errors. These limitations underscore the pressing need for a fully automated, deep-learning-based multi-classification system for brain malignancies. This article aims to leverage a deep convolutional neural network (CNN) to enhance early detection and presents three distinct CNN models designed for different types of classification tasks. The first CNN model achieves an impressive detection accuracy of 99.53% for brain tumors. The second CNN model, with an accuracy of 93.81%, proficiently categorizes brain tumors into five distinct types: normal, glioma, meningioma, pituitary, and metastatic. Furthermore, the third CNN model demonstrates an accuracy of 98.56% in accurately classifying brain tumors into their different grades. To ensure optimal performance, a grid search optimization approach is employed to automatically fine-tune all the relevant hyperparameters of the CNN models. The utilization of large, publicly accessible clinical datasets results in robust and reliable classification outcomes. This article conducts a comprehensive comparison of the proposed models against classical models, such as AlexNet, DenseNet121, ResNet-101, VGG-19, and GoogleNet, reaffirming the superiority of the deep CNN-based approach in advancing the field of brain tumor classification and early detection.

CEACAM7 expression contributes to early events of pancreatic cancer.

BACKGROUND: The trends of pancreatic cancer (PanCa) incidence and mortality are on rising pattern, and it will be a second leading cause of cancer related deaths by 2030. Pancreatic ductal adenocarcinoma (PDAC), major form of PanCa, exhibits a grim prognosis as mortality rate is very close to the incidence rate, due to lack of early detection methods and effective therapeutic regimen. Considering this alarming unmet clinic need, our team has identified a novel oncogenic protein, carcinoembryonic antigen-related cell adhesion molecule 7 (CEACAM7), that can be useful for spotting early events of PDAC. METHODOLOGY: This study includes bioinformatics pre-screening using publicly available cancer databases followed by molecular biology techniques in PDAC progressive cell line panel and human tissues to evaluate CEACAM7 expression in early events of pancreatic cancer. RESULTS: PanCa gene and protein expression analysis demonstrated the significantly higher expression of CEACAM7 in PDAC, compared to other cancers and normal pancreas. Overall survival analysis demonstrated an association between the higher expression of CEACAM7 and poor patients' prognosis with high hazard ratio. Additionally, in a performance comparison analysis CEACAM7 outperformed S100A4 in relation to PDAC. We also observed an increase of CEACAM7 in PDAC cell line panel model. However, poorly differentiated, and normal cell lines did not show any expression. Human tissue analysis also strengthened our data by showing strong and positive IHC staining in early-stage tumors. CONCLUSION: Our observations clearly cite that CEACAM7 can serve as a potential early diagnostic and/or prognostic marker of PDAC and may also potentiate the sensitivity of the existing biomarker panel of PDAC. However, further studies are warranted to determine its clinical significance.

Whole-lesion iodine map histogram analysis versus single-slice spectral CT parameters for determining novel International Association for the Study of Lung Cancer grade of invasive non-mucinous pulmonary adenocarcinomas.

PURPOSE: The purpose of this study was to evaluate and compare the performances of whole-lesion iodine map histogram analysis to those of single-slice spectral computed tomography (CT) parameters in discriminating between low-to-moderate grade invasive non-mucinous pulmonary adenocarcinoma (INMA) and high-grade INMA according to the novel International Association for the Study of Lung Cancer grading system of INMA. MATERIALS AND METHODS: Sixty-one patients with INMA (34 with low-to-moderate grade [i.e., grade I and grade II] and 27 with high grade [i.e., grade III]) were evaluated with spectral CT. There were 28 men and 33 women, with a mean age of 56.4 ± 10.5 (standard deviation) years (range: 29-78 years). The whole-lesion iodine map histogram parameters (mean, standard deviation, variance, skewness, kurtosis, entropy, and 1st, 10th, 25th, 50th, 75th, 90th, and 99th percentile) were measured for each INMA. In other sessions, by placing regions of interest at representative levels of the tumor and normalizing them, spectral CT parameters (iodine concentration and normalized iodine concentration) were obtained. Discriminating capabilities of spectral CT and histogram parameters were assessed and compared using area under the ROC curve (AUC) and logistic regression models. RESULTS: The 1st, 10th, and 25th percentiles of the iodine map histogram analysis, and iodine concentration and normalized iodine concentration of single-slice spectral CT parameters were significantly different between high-grade and low-to-moderate grade INMAs (P < 0.001 to P = 0.002). The 1st percentile of histogram parameters (AUC, 0.84; 95% confidence interval [CI]: 0.73-0.92) and iodine concentration (AUC, 0.78; 95% CI: 0.66-0.88) from single-slice spectral CT parameters had the best performance for discriminating between high-grade and low-to-moderate grade INMAs. At ROC curve analysis no significant differences in AUC were found between histogram parameters (AUC = 0.86; 95% CI: 0.74-0.93) and spectral CT parameters (AUC = 0.81; 95% CI: 0.74-0.93) (P = 0.60). CONCLUSION: Both whole-lesion iodine map histogram analysis and single-slice spectral CT parameters help discriminate between low-to-moderate grade and high-grade INMAs according to the novel International Association for the Study of Lung Cancer grading system, with no differences in diagnostic performances.

Rapid On-Site AI-Assisted Grading for Lung Surgery Based on Optical Coherence Tomography.

The determination of resection extent traditionally relies on the microscopic invasiveness of frozen sections (FSs) and is crucial for surgery of early lung cancer with preoperatively unknown histology. While previous research has shown the value of optical coherence tomography (OCT) for instant lung cancer diagnosis, tumor grading through OCT remains challenging. Therefore, this study proposes an interactive human-machine interface (HMI) that integrates a mobile OCT system, deep learning algorithms, and attention mechanisms. The system is designed to mark the lesion's location on the image smartly and perform tumor grading in real time, potentially facilitating clinical decision making. Twelve patients with a preoperatively unknown tumor but a final diagnosis of adenocarcinoma underwent thoracoscopic resection, and the artificial intelligence (AI)-designed system mentioned above was used to measure fresh specimens. Results were compared to FSs benchmarked on permanent pathologic reports. Current results show better differentiating power among minimally invasive adenocarcinoma (MIA), invasive adenocarcinoma (IA), and normal tissue, with an overall accuracy of 84.9%, compared to 20% for FSs. Additionally, the sensitivity and specificity, the sensitivity and specificity were 89% and 82.7% for MIA and 94% and 80.6% for IA, respectively. The results suggest that this AI system can potentially produce rapid and efficient diagnoses and ultimately improve patient outcomes.

Association of sex-specific abdominal adipose tissue with WHO/ISUP grade in clear cell renal cell carcinoma.

OBJECTIVES: To explore the association between computed tomography (CT)-measured sex-specific abdominal adipose tissue and the pathological grade of clear cell renal cell carcinoma (ccRCC). METHODS: This retrospective study comprised 560 patients (394 males and 166 females) with pathologically proven ccRCC (467 low- and 93 high-grade). Abdominal CT images were used to assess the adipose tissue in the subcutaneous, visceral, and intermuscular regions. Subcutaneous fat index (SFI), visceral fat index (VFI), intermuscular fat index (IFI), total fat index (TFI), and relative visceral adipose tissue (rVAT) were calculated. Univariate and multivariate logistic regression analyses were performed according to sex to identify the associations between fat-related parameters and pathological grade. RESULTS: IFI was significantly higher in high-grade ccRCC patients than in low-grade patients for both men and women. For male patients with high-grade tumors, the SFI, VFI, TFI, and rVAT were significantly lower, but not for female patients. In both univariate and multivariate studies, the IFI continued to be a reliable and independent predictor of high-grade ccRCC, regardless of sex. CONCLUSIONS: Intermuscular fat index proved to be a valuable biomarker for the pathological grade of ccRCC and could be used as a reliable independent predictor of high-grade ccRCC for both males and females. CRITICAL RELEVANCE STATEMENT: Sex-specific fat adipose tissue can be used as a new biomarker to provide a new dimension for renal tumor-related research and may provide new perspectives for personalized tumor management decision-making approaches. KEY POINTS: • There are sex differences in distribution of subcutaneous fat and visceral fat. • The SFI, VFI, TFI, and rVAT were significantly lower in high-grade ccRCC male patients, but not for female patients. • Intermuscular fat index can be used as a reliable independent predictor of high-grade ccRCC for both males and females.

One label is all you need: Interpretable AI-enhanced histopathology for oncology.

Artificial Intelligence (AI)-enhanced histopathology presents unprecedented opportunities to benefit oncology through interpretable methods that require only one overall label per hematoxylin and eosin (H&E) slide with no tissue-level annotations. We present a structured review of these methods organized by their degree of verifiability and by commonly recurring application areas in oncological characterization. First, we discuss morphological markers (tumor presence/absence, metastases, subtypes, grades) in which AI-identified regions of interest (ROIs) within whole slide images (WSIs) verifiably overlap with pathologist-identified ROIs. Second, we discuss molecular markers (gene expression, molecular subtyping) that are not verified via H&E but rather based on overlap with positive regions on adjacent tissue. Third, we discuss genetic markers (mutations, mutational burden, microsatellite instability, chromosomal instability) that current technologies cannot verify if AI methods spatially resolve specific genetic alterations. Fourth, we discuss the direct prediction of survival to which AI-identified histopathological features quantitatively correlate but are nonetheless not mechanistically verifiable. Finally, we discuss in detail several opportunities and challenges for these one-label-per-slide methods within oncology. Opportunities include reducing the cost of research and clinical care, reducing the workload of clinicians, personalized medicine, and unlocking the full potential of histopathology through new imaging-based biomarkers. Current challenges include explainability and interpretability, validation via adjacent tissue sections, reproducibility, data availability, computational needs, data requirements, domain adaptability, external validation, dataset imbalances, and finally commercialization and clinical potential. Ultimately, the relative ease and minimum upfront cost with which relevant data can be collected in addition to the plethora of available AI methods for outcome-driven analysis will surmount these current limitations and achieve the innumerable opportunities associated with AI-driven histopathology for the benefit of oncology.

Relationship between magnetic resonance imaging findings and histological grade in spinal peripheral nerve sheath tumors in dogs.

BACKGROUND: Peripheral nerve sheath tumors (PNSTs) are a group of neoplasms originating from Schwann cells or pluripotent cell of the neural crest. Therapeutic options and prognosis are influenced by their degree of malignancy and location. HYPOTHESIS/OBJECTIVES: Identify magnetic resonance imaging (MRI) features predictive of PNST histologic grade. ANIMALS: Forty-four dogs with histopathological diagnosis of spinal PNSTs and previous MRI investigation. METHODS: A multicenter retrospective study including cases with (a) histopathologic diagnosis of PNST and (b) MRI studies available for review. Histologic slides were reviewed and graded by a board-certified pathologist according to a modified French system (FNCLCC) for grading soft tissue sarcomas. The MRI studies were reviewed by 2 board-certified radiologists blinded to the grade of the tumor and the final decision on the imaging characteristics was reached by consensus. Relationships between tumor grade and histological and MRI findings were assessed using statistical analysis. RESULTS: Forty-four cases met inclusion criteria; 16 patients were PNSTs Grade 1 (low-grade), 19 were PNSTs Grade 2 (medium-grade), and 9 were PNSTs Grade 3 (high-grade). Large volume (P = .03) and severe peripheral contrast enhancement (P = .04) were significantly associated with high tumor grade. Degree of muscle atrophy, heterogeneous signal and tumor growth into the vertebral canal were not associated with grade. CONCLUSIONS AND CLINICAL IMPORTANCE: Grade of malignancy was difficult to identify based on diagnostic imaging alone. However, some MRI features were predictive of high-grade PNSTs including tumor size and peripheral contrast enhancement.

Multinucleated tumor cells in clear cell renal cell carcinoma.

OBJECTIVES: Multinucleated tumor cells (MTCs) in clear cell renal cell carcinoma (ccRCC) are not well understood. METHODS: Our study included ccRCC cases in a single institution between 2010 and 2019. We classified MTC as MTC with degenerative atypia (MTCD), MTC with no anaplasia (MTCNA), and MTC with anaplasia (MTCA). Clinicopathologic characteristics and outcomes were compared between MTC groups. RESULTS: In all, 92 of 256 people (36%) with ccRCC had MTC. People with ccRCC with MTCD and those with ccRCC but no MTC had similar clinicopathologic characteristics and outcomes. Also, MTCNA and MTCA were associated with larger tumor size, advanced pathologic tumor stage, higher World Health Organization/International Society of Urologic Pathologists nuclear grade, and higher metastatic potential (P < .001 for each parameter). Overall, MTCA was associated with an increased rate of recurrence (P = .004), higher metastatic potential (P < .001), and shorter time to metastasis (P = .033), regardless of tumor stage. Univariate Cox regression revealed MTCNA as a significant predictor of metastasis at 5 years (hazard ratio [HR], 4.171; 95% CI, 1.934-8.998); moreover, MTCA was a significant predictor of recurrence (HR, 5.723; 95% CI, 2.495-13.124), metastasis (HR, 12.024; 5.966-24.232), and death (HR, 5.661; 95% CI, 2.688-11.924) at 5 years. CONCLUSIONS: Although MTCD may not be relevant in tumor grading, MTCNA and MTCA are associated with adverse outcomes.

AI-Based Glioma Grading for a Trustworthy Diagnosis: An Analytical Pipeline for Improved Reliability.

Glioma is the most common type of tumor in humans originating in the brain. According to the World Health Organization, gliomas can be graded on a four-stage scale, ranging from the most benign to the most malignant. The grading of these tumors from image information is a far from trivial task for radiologists and one in which they could be assisted by machine-learning-based decision support. However, the machine learning analytical pipeline is also fraught with perils stemming from different sources, such as inadvertent data leakage, adequacy of 2D image sampling, or classifier assessment biases. In this paper, we analyze a glioma database sourced from multiple datasets using a simple classifier, aiming to obtain a reliable tumor grading and, on the way, we provide a few guidelines to ensure such reliability. Our results reveal that by focusing on the tumor region of interest and using data augmentation techniques we significantly enhanced the accuracy and confidence in tumor classifications. Evaluation on an independent test set resulted in an AUC-ROC of 0.932 in the discrimination of low-grade gliomas from high-grade gliomas, and an AUC-ROC of 0.893 in the classification of grades 2, 3, and 4. The study also highlights the importance of providing, beyond generic classification performance, measures of how reliable and trustworthy the model's output is, thus assessing the model's certainty and robustness.

A Weakly Supervised Deep Learning Model and Human-Machine Fusion for Accurate Grading of Renal Cell Carcinoma from Histopathology Slides.

(1) Background: The Fuhrman grading (FG) system is widely used in the management of clear cell renal cell carcinoma (ccRCC). However, it is affected by observer variability and irreproducibility in clinical practice. We aimed to use a deep learning multi-class model called SSL-CLAM to assist in diagnosing the FG status of ccRCC patients using digitized whole slide images (WSIs). (2) Methods: We recruited 504 eligible ccRCC patients from The Cancer Genome Atlas (TCGA) cohort and obtained 708 hematoxylin and eosin-stained WSIs for the development and internal validation of the SSL-CLAM model. Additionally, we obtained 445 WSIs from 188 ccRCC eligible patients in the Clinical Proteomic Tumor Analysis Consortium (CPTAC) cohort as an independent external validation set. A human-machine fusion approach was used to validate the added value of the SSL-CLAM model for pathologists. (3) Results: The SSL-CLAM model successfully diagnosed the five FG statuses (Grade-0, 1, 2, 3, and 4) of ccRCC, and achieved AUCs of 0.917 and 0.887 on the internal and external validation sets, respectively, outperforming a junior pathologist. For the normal/tumor classification (Grade-0, Grade-1/2/3/4) task, the SSL-CLAM model yielded AUCs close to 1 on both the internal and external validation sets. The SSL-CLAM model achieved a better performance for the two-tiered FG (Grade-0, Grade-1/2, and Grade-3/4) task, with AUCs of 0.936 and 0.915 on the internal and external validation sets, respectively. The human-machine diagnostic performance was superior to that of the SSL-CLAM model, showing promising prospects. In addition, the high-attention regions of the SSL-CLAM model showed that with an increasing FG status, the cell nuclei in the tumor region become larger, with irregular contours and increased cellular pleomorphism. (4) Conclusions: Our findings support the feasibility of using deep learning and human-machine fusion methods for FG classification on WSIs from ccRCC patients, which may assist pathologists in making diagnostic decisions.

Coordinatized lesion location analysis empowering ROI-based radiomics diagnosis on brain gliomas.

OBJECTIVES: To assess the value of coordinatized lesion location analysis (CLLA), in empowering ROI-based imaging diagnosis of gliomas by improving accuracy and generalization performances. METHODS: In this retrospective study, pre-operative contrasted T1-weighted and T2-weighted MR images were obtained from patients with gliomas from three centers: Jinling Hospital, Tiantan Hospital, and the Cancer Genome Atlas Program. Based on CLLA and ROI-based radiomic analyses, a fusion location-radiomics model was constructed to predict tumor grades, isocitrate dehydrogenase (IDH) status, and overall survival (OS). An inter-site cross-validation strategy was used for assessing the performances of the fusion model on accuracy and generalization with the value of area under the curve (AUC) and delta accuracy (ACC) (ACCtesting-ACCtraining). Comparisons of diagnostic performances were performed between the fusion model and the other two models constructed with location and radiomics analysis using DeLong's test and Wilcoxon signed ranks test. RESULTS: A total of 679 patients (mean age, 50 years ± 14 [standard deviation]; 388 men) were enrolled. Based on tumor location probabilistic maps, fusion location-radiomics models (averaged AUC values of grade/IDH/OS: 0.756/0.748/0.768) showed the highest accuracy in contrast to radiomics models (0.731/0.686/0.716) and location models (0.706/0.712/0.740). Notably, fusion models ([median Delta ACC: - 0.125, interquartile range: 0.130]) demonstrated improved generalization than that of radiomics model ([- 0.200, 0.195], p = 0.018). CONCLUSIONS: CLLA could empower ROI-based radiomics diagnosis of gliomas by improving the accuracy and generalization of the models. CLINICAL RELEVANCE STATEMENT: This study proposed a coordinatized lesion location analysis for glioma diagnosis, which could improve the performances of the conventional ROI-based radiomics model in accuracy and generalization. KEY POINTS: • Using coordinatized lesion location analysis, we mapped anatomic distribution patterns of gliomas with specific pathological and clinical features and constructed glioma prediction models. • We integrated coordinatized lesion location analysis into ROI-based analysis of radiomics to propose new fusion location-radiomics models. • Fusion location-radiomics models, with the advantages of being less influenced by variabilities, improved accuracy, and generalization performances of ROI-based radiomics models on predicting the diagnosis of gliomas.

Association of Membranous WNT-1 and Nuclear mTOR with Endometrial Cancer Grade.

Endometrial cancer remains a common cancer affecting the female reproductive system. There is still a need for more efficient ways of determining the degree of malignancy and optimizing treatment. WNT and mTOR are components of signaling pathways within tumor cells, and dysfunction of either protein is associated with the pathogenesis of neoplasms. Therefore, the aim of our study was to assess the impact of subcellular WNT-1 and mTOR levels on the clinical course of endometrial cancer. WNT-1 and mTOR levels in the plasma membrane, nucleus, and cytoplasm were evaluated using immunohistochemical staining in a group of 64 patients with endometrial cancer of grades 1-3 and FIGO stages I-IV. We discovered that the levels of WNT-1 and mTOR expression in the cellular compartments were associated with tumor grade and staging. Membranous WNT-1 was negatively associated, whereas cytoplasmic WNT-1 and nuclear mTOR were positively associated with higher grading of endometrial cancer. Furthermore, nuclear mTOR was positively associated with FIGO stages IB-IV. To conclude, we found that the assessment of WNT-1 in the cell membrane may be useful for exclusion of grade 3 neoplasms, whereas cytoplasmic WNT-1 and nuclear mTOR may be used as indicators for confirmation of grade 3 neoplasms.