Pathological Classification Research Articles

Multi-parametric MRI-based machine learning model for prediction of pathological grade of renal injury in a rat kidney cold ischemia-reperfusion injury model

BackgroundRenal cold ischemia-reperfusion injury (CIRI), a pathological process during kidney transplantation, may result in delayed graft function and negatively impact graft survival and function. There is a lack of an accurate and non-invasive tool for evaluating the degree of CIRI. Multi-parametric MRI has been widely used to detect and evaluate kidney injury. The machine learning algorithms introduced the opportunity to combine biomarkers from different MRI metrics into a single classifier.ObjectiveTo evaluate the performance of multi-parametric magnetic resonance imaging for grading renal injury in a rat model of renal cold ischemia-reperfusion injury using a machine learning approach.MethodsEighty male SD rats were selected to establish a renal cold ischemia -reperfusion model, and all performed multiparametric MRI scans (DWI, IVIM, DKI, BOLD, T1mapping and ASL), followed by pathological analysis. A total of 25 parameters of renal cortex and medulla were analyzed as features. The pathology scores were divided into 3 groups using K-means clustering method. Lasso regression was applied for the initial selecting of features. The optimal features and the best techniques for pathological grading were obtained. Multiple classifiers were used to construct models to evaluate the predictive value for pathology grading.ResultsAll rats were categorized into mild, moderate, and severe injury group according the pathologic scores. The 8 features that correlated better with the pathologic classification were medullary and cortical Dp, cortical T2*, cortical Fp, medullary T2*, ∆T1, cortical RBF, medullary T1. The accuracy(0.83, 0.850, 0.81, respectively) and AUC (0.95, 0.93, 0.90, respectively) for pathologic classification of the logistic regression, SVM, and RF are significantly higher than other classifiers. For the logistic model and combining logistic, RF and SVM model of different techniques for pathology grading, the stable and perform are both well. Based on logistic regression, IVIM has the highest AUC (0.93) for pathological grading, followed by BOLD(0.90).ConclusionThe multi-parametric MRI-based machine learning model could be valuable for noninvasive assessment of the degree of renal injury.

Characteristics of myopic maculopathy in Chinese children and adolescents with high myopia

AimsTo investigate the characteristics of myopic maculopathy among highly myopic Chinese children and adolescents and explore its associated risk factors.MethodsChildren and adolescents aged 7–17 years with spherical equivalent (SE) ≤...

Urinary renal tubular epithelial cells and casts as predictors of renal outcomes in patients with biopsy-proven diabetic nephropathy.

Urine sediment examination is a time-tested and non-invasive diagnostic tool. Thisstudy investigated the characteristics of urine sediment and its association with severity and renal outcomes in diabetic nephropathy (DN) patients. A total of 201 biopsy-proven diabetic nephropathy patients (according to the pathological classification of diabetic nephropathyproposed by the Renal Pathology Society in 2010)who underwent manual urine sediment microscopic examination were included. We compared the clinicopathologicalcharacteristics of diabetic nephropathy patients with and without urinary renal tubular epithelial cells (RTECs) or renal tubular epithelial cell casts. The predictive value of urinary renal tubular epithelial cells or renal tubular epithelial cell casts for renal outcomes in diabetic nephropathy was analyzed. Fifty of 201 (24.9%) diabetic nephropathy patients had renal tubular epithelial cells or renal tubular epithelial cell casts in urine sediment. Diabetic nephropathy patients with renal tubular epithelial cells or renal tubular epithelial cell casts in urine sediment had a significantly higher level of proteinuria [6.0 (3.1, 9.7)vs.3.6 (1.8, 6.8) g/24h, p = 0.001], higher serum creatinine [227.9 (151.6, 338.1)vs.177.0 (104.4, 288.4) μmol/L, p = 0.016] and lower estimated glomerular filtration rate (eGFR) [25.8 (15.8, 44.8)vs.35.7 (19.9, 65.0) mL/min/1.73m2, p = 0.025] than those without.Cox regression analysis demonstrated that the presence of urinaryrenal tubular epithelial cells or renal tubular epithelial cell casts was independently associated with the development of end-stage kidney disease (ESKD) in diabetic nephropathy patients [HR 1.670, 95% CI (1.042, 2.676), p = 0.033]. Adding the presence of urinary renal tubular epithelial cells or renal tubular epithelial cell casts to the predictive model could improve the effectiveness of the model for predicting the risk of ESKD within one year after renal biopsy. The presence of urinary renal tubular epithelial cells or renal tubular epithelial cell casts was associated with more severe kidney injury and worse renal outcomes in patients with diabetic nephropathy, thus perhaps providing a noninvasive biomarker for predicting diabetic nephropathy.

Distinct phenotype and prognosis of immune-mediated necrotizing myopathy based on clinical-serological-pathological classification.

The aim of the study was to investigate the characteristics and prognosis of patients with immune-mediated necrotizing myopathy (IMNM) based on clinical, serological and pathological classification. A total of 138 patients with IMNM who met the 2018 European Neuromuscular Center criteria for IMNM including 62 anti-SRP, 32 anti-HMGCR-positive and 44 myositis specific antibody-negative were involved in the study. All patients were followed up and evaluated remission and relapse. Clustering analysis based on clinical, serological, and pathological parameters was used to define subgroups. Clustering analysis classified IMNM into three clusters. Cluster 1 patients (n = 35) had the highest CK levels, the shortest disease course, severe muscle weakness, and more inflammation infiltration in muscle biopsy. Cluster 2 patients (n = 79) had the lowest CK level and moderate inflammation infiltrate. Cluster 3 patients (n = 24) had the youngest age of onset, the longest disease course and the least frequency of inflammatory infiltration. Patients in cluster 3 had the longest time-to-remission (median survival time: 61[18.3, 103.7] vs 20.5[16.2, 24.9] and 27[19.6, 34.3] months) and shortest relapse-free time than those in cluster 1 and 2 (median remission time 95%CI: 34[19.9, 48.0] vs 73[49.0, 68.7] and 73[48.4, 97.6] months). Patients with age of onset >55 years, more regeneration of muscle fibers, more CD4+T infiltration, and MAC deposition had more favorable outcomes regarding time to achieving remission. Stratification combining clinical, serological, and pathological features could distinguish phenotypes and prognosis of IMNM. The pathological characteristics may impact the long-term prognosis of patients with IMNM.

Low-Grade Urothelial Neoplasia

This chapter reviews the clinical, histopathologic, and cytologic features of low-grade urothelial neoplasia, including urothelial papilloma, papillary urothelial neoplasm of low malignant potential, and low-grade papillary urothelial carcinomas. An overview of various surgical pathology and cytology classification systems, diagnostic criteria, and common molecular alterations will be reviewed.

DNA methylation analysis in plasma for early diagnosis in lung adenocarcinoma.

Lung adenocarcinoma (LUAD) represents the most prevalent type of lung cancer. SHOX2 and RASSF1A methylation have been identified as important biomarkers for diagnosis and prognosis of lung cancer. Bronchoalveolar lavage fluid (BALF) exhibits good specificity and sensitivity in diagnosing pulmonary diseases, but its acquisition is challenging and may cause discomfort to patients. In clinical, plasma samples are more convenient to obtain than BALF; however, there is little research on the concurrent detection of SHOX2 and RASSF1A methylation in plasma. This study aims to assess the diagnostic value of a combined promoter methylation assay for SHOX2 and RASSF1A in early-stage LUAD using plasma samples. BALF and blood samples were obtained from 36 early-stage LUAD patients, with a control group of nineteen non-tumor individuals. The promoter methylation levels of SHOX2 and RASSF1A in all subjects were assessed using the human SHOX2 and RASSF1A gene methylation kit. The methylation detection rate of SHOX2 and RASSF1A in plasma was 61.11%, slightly lower than that in BALF (66.7%). The Chi-square test revealed no significant difference in the methylation rate between BALF and plasma (P > 0.05). The area under the receiver operating characteristic (ROC) curve analysis for blood was 0.806 (95% CI, 0.677 to 0.900), while for BALF it was 0.781 (95% CI, 0.649 to 0.881). Additionally, we conducted an analysis on the correlation between SHOX2 and RASSF1A methylation levels in plasma with gender, age, tumor differentiation, pathologic classification, and other clinicopathological variables; however, no significant correlations were observed. Measurement of SHOX2 and RASSF1A methylation for early diagnosis of LUAD can be achieved with high sensitivity and specificity by using plasma as a substitute for BALF samples.

A Comprehensive Framework for Pathology Classification Bridging Precision and Interpretability

Pathology classification is an indispensable component of medical diagnostics, facilitating accurate disease identification, prognosis determination, and treatment planning. However, the increasing complexity and heterogeneity of pathological manifestations pose significant challenges to traditional classification methodologies. This abstract presents a novel framework that integrates advanced machine learning techniques with domain-specific expertise to enhance the precision and interpretability of pathology classification. Our framework adopts a multi-modal approach, leveraging diverse data sources including histopathological images, clinical records, genomic profiles, and molecular biomarkers. Through feature fusion and dimensionality reduction techniques, we effectively capture intricate patterns and latent relationships embedded within the data, enabling robust classification across diverse pathological conditions. Furthermore, interpretability is prioritized through the incorporation of explainable AI methodologies, facilitating the identification of salient features and decision rationales underlying classification outcomes. This ensures transparency and trustworthiness in the diagnostic process, empowering clinicians to make informed decisions and refine treatment strategies. Validation of our framework across various pathological contexts demonstrates superior performance compared to conventional approaches, exhibiting high accuracy, sensitivity, and specificity. Moreover, its modular architecture facilitates customization and scalability, accommodating evolving diagnostic needs and emerging technological advancements. In conclusion, our proposed framework represents a significant advancement in pathology classification, offering a synergistic blend of computational sophistication and clinical relevance. By seamlessly integrating cutting-edge technologies with domain knowledge, it holds promise for revolutionizing diagnostic practices and improving patient outcomes in the realm of precision medicine.

Correlations between plasma markers and brain Aβ deposition across the AD continuum: Evidence from SILCODE.

Previous studies have found that Alzheimer's disease (AD)-related plasma markers are associated with amyloid beta (Aβ) deposition, but the change of this association in different Aβ pathological stages remains unclear. Data were obtained from the SILCODE. According to the standardized uptake value ratio (SUVR) and Aβ stage classification, correlation analysis was performed among plasma biomarkers, and voxel/SUVR values in the regions of interest (ROI) and clinical scale information, respectively. Mediation analysis was used to study the possible pathways. The proportion of cognitively normal (CN) and subjective cognitive decline (SCD) was the highest in stages A0 to 1, while in stages A2 to 4, the proportion of mild cognitive impairment (MCI) and AD increased. Plasma phosphorylated tau (p-tau)181 and glial fibrillary acidic protein (GFAP) levels were significantly lower in stage A0 compared to the later phases. Two pathways demonstrated fully mediated effects: positron emission tomography (PET) SUVR-plasma p-tau181-Mini-Mental State Examination (MMSE) and PET SUVR-plasma GFAP-MMSE. This study demonstrated the role of plasma biomarkers in the early stage of AD, especially in SCD, from both the clinical diagnosis and Aβ stage dimensions. Plasma ptau181 and GFAP level serve as indicators of early Alzheimer's disease and the pathologic Aβ staging classification. A possible ceiling effect of GFAP was observed in the mid-to-late stages of the AD course. This study confirms the role of AD plasma markers in promoting Aβ deposition at an early stage, particularly in females with subjective cognitive decline(SCD). The overlapping brain regions of plasma p-tau181, GFAP, and neurofilament light for Aβ deposition in the brain in early AD were distributed across various regions, including the posterior cingulate gyrus, rectus gyrus, and inferior temporal gyrus.

Fast Learning Network Algorithm for Voice Pathology Detection and Classification

Fast Learning Network Algorithm for Voice Pathology Detection and Classification

Long-term outcomes of laparoscopic versus open distal gastrectomy for patients with advanced gastric cancer in North China: a multicenter randomized controlled trial

BackgroundLaparoscopic distal gastrectomy (LDG) has become a common procedure for treating advanced gastric cancer (AGC) in China. However, there is uncertainty regarding its oncological outcomes compared to open distal gastrectomy (ODG). This study aims to compare the 3-year disease-free survival (DFS) rates among patients who underwent surgery for AGC in northern China.MethodsA multicenter, non-inferiority, open-label, parallel, randomized clinical trial was conducted to evaluate patients with AGC who were eligible for distal gastrectomy at five tertiary hospitals in North China. In this trial, patients were randomly assigned preoperatively to receive either LDG or ODG in a 1:1 allocation ratio. The primary endpoint was postoperative morbidity and mortality within 30 days and the secondary endpoint was the 3-year DFS rate. This trial has been registered at ClinicalTrials.gov (Identifier: NCT02464215).ResultsA total of 446 patients were randomly allocated to LDG (n = 223) or ODG group (n = 223) between March 2014 and August 2017. After screening, a total of 214 patients underwent the open surgical approach, while 216 patients underwent laparoscopic surgery. The 3-year DFS rate was 85.9% for the LDG group and 84.72% for the ODG group, with no significant statistical difference (Hazard ratio 1.12; 95% CI 0.68–1.84, P = 0.65). Body mass index (BMI) < 25 kg/m2, advanced pathologic T4, and pathologic N2-3 category were confirmed as independent risk factors for DFS in the Cox regression.ConclusionsIn comparison to ODG, LDG with D2 lymphadenectomy yielded similar outcomes in terms of 3-year DFS rates among patients diagnosed with AGC.

Brain age revisited: Investigating the state vs. trait hypotheses of EEG-derived brain-age dynamics with deep learning

Abstract The brain’s biological age has been considered as a promising candidate for a neurologically significant biomarker. However, recent results based on longitudinal magnetic resonance imaging (MRI) data have raised questions on its interpretation. A central question is whether an increased biological age of the brain is indicative of brain pathology and if changes in brain age correlate with diagnosed pathology (state hypothesis). Alternatively, could the discrepancy in brain age be a stable characteristic unique to each individual (trait hypothesis)? To address this question, we present a comprehensive study on brain aging based on clinical Electroencephalography (EEG), which is complementary to previous MRI-based investigations. We apply a state-of-the-art temporal convolutional network (TCN) to the task of age regression. We train on recordings of the Temple University Hospital EEG Corpus (TUEG) explicitly labeled as non-pathological and evaluate on recordings of subjects with non-pathological as well as pathological recordings, both with examinations at a single point in time TUH Abnormal EEG Corpus (TUAB) and repeated examinations over time. Therefore, we created four novel subsets of TUEG that include subjects with multiple recordings: repeated non-pathological (RNP): all labeled non-pathological; repeated pathological (RP): all labeled pathological; transition non-patholoigical pathological (TNPP): at least one recording labeled non-pathological followed by at least one recording labeled pathological; and transition pathological non-pathological (TPNP): similar to TNPP but with opposing transition (first pathological and then non-pathological). The results show that our TCN reaches state-of-the-art performance in age decoding on non-pathological subjects of TUAB with a mean absolute error of 6.6 years and an R2 score of 0.73. Our extensive analyses demonstrate that the model underestimates the age of non-pathological and pathological subjects, the latter significantly (-1 and -5 years, paired t-test, p = 0.18 and p = 6.6e−3). Furthermore, there exist significant differences in average brain age gap between non-pathological and pathological subjects both with single examinations (TUAB) and repeated examinations (RNP vs. RP) (-4 and -7.48 years, permutation test, p = 1.63e−2 and p = 1e−5). We find mixed results regarding the significance of pathology classification based on the brain age gap biomarker. While it is indicative of pathological EEG in datasets TUAB and RNP versus RP (61.12% and 60.80% BACC, permutation test, p = 1.32e−3 and p = 1e−5), it is not indicative in TNPP and TPNP (44.74% and 47.79% BACC, permutation test, p = 0.086 and p = 0.483). Additionally, all of these classification scores are clearly inferior to the ones obtained from direct EEG pathology classification at 86% BACC and higher. Furthermore, we could not find evidence that a change of EEG pathology status within subjects relates to a significant change in brain age gap in datasets TNPP and TPNP (0.46 and 1.35 years, permutation test, p = 0.825 and p = 0.43; and Wilcoxon-Mann-Whitney and Brunner-Munzel test, p = 0.13). Our findings, thus, support the trait rather than the state hypothesis for brain age estimates derived from EEG. In summary, our findings indicate that the neural underpinnings of brain age changes are likely more multifaceted than previously thought, and that taking this into account will benefit the interpretation of empirically observed brain age dynamics.

Classification and Analysis of HIV Neurocognitive MRI Images using Support Vector Machine

Medical imaging has expanded thanks to advances in processing power and advanced image analysis techniques, especially with magnetic resonance imaging (MRI), which offers comprehensive body scans for diagnosis. This work proposes a simple yet efficient method to use a support vector machine (SVM) to classify HIV neurocognitive MRI pictures into normal and pathological categories. The model consists of four steps: data pre-processing, feature extraction, SVM classification, and model evaluation. To separate desired and undesired elements, such as the scalp and skull, pre-processed images were converted from grayscale to colour using support vector machines. The discrete wavelet transform (DWT) was used in the feature extraction stage to extract image properties. Colour moments (CMs) were then used to optimize the feature collection. Afterwards, the SVM classifier was used to determine the ideal feature set to classify images. For example, a dataset is used for training and testing, with a split ratio of 75% to 25% respectively. The experimental results show that the proposed model has a high classification accuracy of 94.4%

Predictors of lung cancer subtypes and lymph node status in non-small-cell lung cancer: intravoxel incoherent motion parameters and extracellular volume fraction

ObjectiveTo determine the performance of intravoxel incoherent motion (IVIM) parameters and the extracellular volume fraction (ECV) in distinguishing between different subtypes of lung cancer and predicting lymph node metastasis (LNM) status in patients with non-small-cell lung cancer (NSCLC).MethodsOne hundred sixteen patients with lung cancer were prospectively recruited. IVIM, native, and postcontrast T1 mapping examinations were performed, and the T1 values were measured to calculate the ECV. The differences in IVIM parameters and ECV were compared between NSCLC and small-cell lung cancer (SCLC), adenocarcinoma (Adeno-Ca) and squamous cell carcinoma (SCC), and NSCLC without and with LNM. The assessment of each parameter’s diagnostic performance was based on the area under the receiver operating characteristic curve (AUC).ResultsThe apparent diffusion coefficient (ADC), true diffusion coefficient (D), and ECV values in SCLC were considerably lower compared with NSCLC (all p < 0.001, AUC > 0.887). The D value in SCC was substantially lower compared with Adeno-Ca (p < 0.001, AUC = 0.735). The perfusion fraction (f) and ECV values in LNM patients were markedly higher compared with those without LNM patients (p < 0.01, < 0.001, AUC > 0.708).ConclusionIVIM parameters and ECV can serve as non-invasive biomarkers for assisting in the pathological classification and LNM status assessment of lung cancer patients.Critical relevance statementIVIM parameters and ECV demonstrated remarkable potential in distinguishing pulmonary carcinoma subtypes and predicting LNM status in NSCLC.Key PointsLung cancer is prevalent and differentiating subtype and invasiveness determine the treatment course.True diffusion coefficient and ECV showed promise for subtyping and determining lymph node status.These parameters could serve as non-invasive biomarkers to help determine personalized treatment strategies.Graphical

Comparison of the diagnostic efficacy of systemic inflammatory indicators in the early diagnosis of ovarian cancer.

This study aimed to determine the diagnostic accuracy of CA125, HE4, systemic immune-inflammation index (SII), prognostic nutritional index (PNI), fibrinogen-to-albumin ratio (FAR), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio (MLR), and the combination of the six inflammatory-nutritional markers for ovarian cancer (OC) to identify the best diagnostic indicator for OC early diagnosis. An extensive study was performed to establish the connection between these indicators and the pathological aspects of OC. A total of 170 individuals were included in this study, with 87 diagnosed with OC and 83 with benign ovarian tumors (BOTs). The diagnostic abilities of the variables were evaluated by calculating sensitivity, specificity, and area under the ROC curves. Through the use of DCA, we evaluated the variables' clinical value in the discrimination of ovarian masses. All markers showed significant diagnostic power for OC. CA125, HE4, SII, FAR, and MLR levels significantly increased from the BOTs group to the early-stage OC group. The advanced-stage OC group had significantly lower PNI values compared to the early-stage OC group but significantly higher levels of CA125, HE4, SII, NLR, and FAR. Moreover, the OC group with lymph node metastasis exhibited significantly higher levels of CA125, HE4, SII, NLR, PLR, and FAR, in contrast to the non-metastatic group, while PNI levels were significantly lower. Categorical factors, such as histological grade and pathological classification, showed noticeable discrepancies in CA125 and HE4 levels. NLR was significantly different among the pathological type groups. Among the six inflammatory-nutritional markers, the FAR displayed the greatest diagnostic value. In the analysis of logistic regression, it was observed that a combination marker containing all six inflammatory-nutritional markers exhibited a notably higher AUC value (0.881; 95% CI, 0.823 - 0.926) than any of the individual marker. PNI, NLR, PLR, MLR, SII, and FAR showed excellent diagnostic performance for OC. The combination of these markers demonstrated a superior diagnostic capability compared to each individual one. The systemic inflammatory indicators may be helpful to diagnose OC.

A lightweight spatially-aware classification model for breast cancer pathology images

Breast cancer is a prevalent malignant tumour with high global incidence. Its diagnosis relies primarily on the analysis of pathological breast images. Owing to the complex organisation of the tumour microenvironment, neural network models are essential as efficient classification tools in the field of pathological image analysis. This study introduced spatially-aware attention swift parallel convolution network (SPA-SPCNet), a lightweight and low-latency model for classifying breast pathologies. A novel module for multi-scale feature extraction was constructed using a depthwise separable convolution method. It focuses on the multi-scale features of pathological images to alleviate recognition problems caused by similar local features in breast cancer tissues. The module concatenates the convolutions of different kernels from three branches. Second, a lightweight dynamic spatially-aware attention module was introduced to integrate the visual graph convolutional architecture in a branch. This allowed the model to capture the spatial structure and relationships in image, enabling better handling of the unique spatial distribution relationship between breast cancer tissue structures. The other branch utilises a self-attention mechanism in the transformer. The module can dynamically adjust the attention of the model to different regions in the image, allowing it to focus on the key features of the complex spatial distribution of breast cancer tissue. This feature fusion method enabled the model to capture both global semantics and local details. Compared with existing lightweight models, the proposed model has advantages in terms of tissue structure classification accuracy, parameter quantity, floating-point operations, and real-time inference speed, providing a powerful tool for computer-aided breast pathological image classification.

Diagnostic accuracy of cerebroplacental ratio in anticipating adverse perinatal outcome in uncomplicated appropriate for gestational age pregnancies at term

Introduction: Anticipating which babies are in danger of experiencing poor outcomes during the perinatal period in uncomplicated appropriate for gestational age (AGA) pregnancies at term is difficult in obstetric practice. Cerebroplacental ratio (CPR) is emerging as a significant indicator of negative perinatal results. The current study sought to establish how effective CPR is in predicting negative perinatal outcomes in term uncomplicated AGA pregnancies. Materials and Methods: This study was a hospital-based prospective observational cohort study conducted at a single center. Patients were chosen depending on different criteria for inclusion and exclusion. A prenatal ultrasound with a color Doppler scan was carried out to calculate CPR. Patients were grouped according to their last CPR measurement before delivery into either normal CPR or pathological CPR categories. Doppler results did not impact clinical decisions and delivery followed institutional protocols. After childbirth, data on the outcome of the perinatal period were obtained from the patients’ medical records. Negative perinatal outcomes were assessed through the delivery method, APGAR score, perinatal morbidity, and perinatal mortality. These outcomes were correlated with CPR. Results: The research included 605 women, separated into normal and pathological CPR groups. Of these, 153 (25.3%) were classified in the pathological CPR category, whereas 452 (74.7%) exhibited normal CPR. In our study, there were 138 patients who experienced adverse perinatal outcomes in the pathological CPR group, whereas 44 patients had such outcomes in the normal CPR group. The diagnostic accuracy of pathological CPR to predict any negative perinatal result was 90.25%. Conclusions: CPR shows potential in detecting fetuses at risk in full-term uncomplicated AGA pregnancies.

Pathological voice classification using MEEL features and SVM-TabNet model

In clinical settings, early diagnosis and objective assessment depend on the detection of voice pathology. To classify anomalous voices, this work uses an approach that combines the SVM-TabNet fusion model with MEEL (Mel-Frequency Energy Line) features. Further, the dataset consists of 1037 speech files, including recordings from people with laryngocele and Vox senilis as well as from healthy persons. Additionally, the main goal is to create an efficient classification model that can differentiate between normal and abnormal voice patterns. Modern techniques frequently lack the accuracy required for a precise diagnosis, which highlights the need for novel strategies. The suggested approach uses an SVM-TabNet fusion model for classification after feature extraction using MEEL characteristics. MEEL features provide extensive information for categorization by capturing complex patterns in audio transmissions. Moreover, by combining the advantages of SVM and TabNet models, classification performance is improved. Moreover, testing the model on test data yields remarkable results: 99.7 % accuracy, 0.992 F1 score, 0.996 precision, and 0.995 recall. Additional testing on additional datasets reliably validates outstanding performance, with 99.4 % accuracy, 0.99 F1 score, 0.998 precision, and 0.989 % recall. Furthermore, using the Saarbruecken Voice Database (SVD), the suggested methodology achieves an impressive accuracy of 99.97 %, demonstrating its durability and generalizability across many datasets. Overall, this work shows how the SVM-TabNet fusion model with MEEL characteristics may be used to accurately and consistently classify diseased voices, providing encouraging opportunities for clinical diagnosis and therapy tracking.

Tumor budding -a potential biomarker in low grade salivary gland carcinomas?

Low-grade salivary gland carcinoma is regularly treated with surgical therapy of the salivary gland without elective neck dissection in T1/2 carcinomas, either alone or with adjuvant radiation therapy. However, occult metastasis and locoregional recurrence influence therapy and outcome. Tumor budding is an emerging prognostic pathological factor in many carcinomas, but has not yet been adequately considered in salivary gland carcinomas. We conducted a retrospective single-center study of 64 patients diagnosed with low-grade carcinoma of the major salivary glands treated between 2003 and 2017. Pathological risk factors and TNM classification were thoroughly assessed for each case. All hematoxylin and eosin (HE)-stained histological specimens underwent careful examination, and tumor budding was identified following the guidelines set forth by the International Tumor Budding Consensus Conference in 2016. Tumor budding was not statistically significant concerning 5-year survival rate (5-YSR) (p=0.969) and mean overall survival (log-rank p=0.315). Whereas 5-year disease-free survival rate (5-YDFSR) was 87% in the low tumor budding group and 61.1% in the intermediate and high tumor budding group (p=0.021). Mean disease-free survival accounted for 100.2 months (CI: 88.6;111.9) in the low budding score group and 58.7 months (CI: 42.8;74.6) in the other group (log-rank p=0.032). Notably, pT1/2 showed significantly lower tumor buds than pT3/4 stages (2.43 tumor buds/0.785 mm2 vs. 4.19 tumor buds/0.785 mm2, p=0.034). Similar findings were noted comparing nodal-positive and nodal-negative patients, as well as patients with and without lymphovascular invasion and perineural invasion (each p<0.05). Tumor budding might be used as an additional prognostic factor for recurrence in low-grade salivary gland carcinoma, seemingly associated with a higher nodal metastasis rate and advanced tumor stages and a worse 5-YDFSR. Consequently, the evaluation of tumor budding in resection specimens of low-grade salivary gland tumor may prove valuable in decision-making for neck dissection and follow-up strategy.

Contrastive Learning Joint Regularization for Pathological Image Classification with Noisy Labels

The annotation of pathological images often introduces label noise, which can lead to overfitting and notably degrade performance. Recent studies have attempted to address this by filtering samples based on the memorization effects of DNNs. However, these methods often require prior knowledge of the noise rate or a small, clean validation subset, which is extremely difficult to obtain in real medical diagnosis processes. To reduce the effect of noisy labels, we propose a novel training strategy that enhances noise robustness without prior conditions. Specifically, our approach includes self-supervised regularization to encourage the model to focus more on the intrinsic connections between images rather than relying solely on labels. Additionally, we employ a historical prediction penalty module to ensure consistency between successive predictions, thereby slowing down the model’s shift from memorizing clean labels to memorizing noisy labels. Furthermore, we design an adaptive separation module to perform implicit sample selection and flip the labels of noisy samples identified by this module and mitigate the impact of noisy labels. Comprehensive evaluations of synthetic and real pathological datasets with varied noise levels confirm that our method outperforms state-of-the-art methods. Notably, our noise handling process does not require any prior conditions. Our method achieves highly competitive performance in low-noise scenarios which aligns with current pathological image noise situations, showcasing its potential for practical clinical applications.

The value of the radiological diameter-to-thickness ratio in patients with HER2-positive resectable advanced gastric cancer: implications for long survival and stage migration.

To investigate the clinical significance and stage migration effect of radiological diameter-to-thickness (DT) ratio in HER2-positive resectable advanced gastric cancer (HER2-p RAGC). 369 HER2-p RAGC patients were retrospectively enrolled and information on clinical pathological characteristics, radiological DT ratio, and outcomes [i.e., overall survival (OS) and progression-free survival (PFS)] was collected. Pearson's Chi-square and Student's t-test were employed to compare baseline characteristics. Clinical outcomes were estimated using Kaplan-Meier analysis and Log-rank test. Univariate and multivariate Cox regression models were utilized to analyze independent prognostic factors. HER2-p RAGC patients were stratified into two groups using a DT ratio cutoff value of 4.0 (p < 0.05). Patients with a DT ratio < 4.0 exhibited significantly longer OS (58.0 vs. 31.0months) and PFS (43.0 vs. 24.0months) than those with a DT ratio ≥ 4.0. DT ratio significantly predicted prognosis for N0 and II stage patients (p < 0.05). Patients with gastric body and antrum cancers demonstrated longer OS and PFS in the DT ratio < 4.0 group (p = 0.046, 0.017, 0.036 and 0.028). Multivariate Cox proportional hazard model identified age, pathological T category, pathological N category, pathological TNM category and DT ratio as independent prognostic factors. Notably, pStage II patients with a DT ratio ≥ 4.0 exhibited a similar prognosis to pStage III patients with a DT ratio < 4.0 (p = 0.418 for OS, 0.867 for PFS). Radiological DT ratio could evaluate the prognosis and detect higher malignant cases in HER2-p RAGC patients. Moreover, DT ratio might guide clinicians make postoperative strategies. Retrospectively registered.