Added Value of Histological Evaluation of Muscle Biopsies in Porcine Vascularized Composite Allografts.

Background: Machine perfusion (MP) offers extended preservation of vascularized complex allografts (VCA), but the diagnostic value of histology using hematoxylin and eosin (H&E) in detecting ischemia-reperfusion injury (IRI) in muscle cells remains unclear. This study aims to document the application of the Histology Injury Severity Score (HISS) and to assess whether additional staining for nicotinamide adenine dinucleotide (NADH) and membrane attack complex (MAC) improves IRI detection in a porcine limb replantation model. Methods: The forelimbs of 16 Dutch Landrace pigs were amputated and preserved for 24 h using hypothermic MP (n = 8) with Histidine-Tryptophan-Ketoglutarate (HTK) or for 4 h with SCS (n = 8) before heterotopic replantation and 7 days of follow-up. Muscle damage was assessed via biochemical markers and light microscopy using H&E, NADH, and MAC at baseline, post-intervention, and post-operative day (POD) 1, 3, and 7 timepoints, using the HISS and a self-developed NADH and MAC score. Results: H&E effectively identified damaged muscle fibers and contributed to IRI assessment in porcine limbs (p < 0.05). The highest HISS was measured on POD 3 between MP (4.9) and SCS (3.5) (p = 0.029). NADH scores of both preservation groups varied over the 7-day follow-up and were statistically insignificant compared with baseline measurements (p > 0.05). MAC revealed no to minimal necrotic tissue across the different timepoints. Conclusions: This study documents the application of the HISS with H&E to detect IRI in muscle fibers. NADH and MAC showed no significant added diagnostic utility. The 24 h MP showed similar muscle alterations using the HISS compared to that of the 4 h SCS after a 7-day follow up.

Evaluation of Safe Margins Using Immunohistochemistry in Oral Squamous Cell Carcinoma.

Background: Hematoxylin & Eosin (H & E) stains have been conventionally used to establish the status of safe margins following resection of primary Oral Squamous Cell Carcinoma. Due to non-specificity of this stain, there is a possibility of false negative results. In this study, we have assessed the role of Immunohistochemistry (IHC) in establishing the status of safe margins. Aim: To compare Hematoxylin & Eosin (H & E) and Immunohistochemistry (IHC) staining in identification of tumor cells in establishing the status of safe margins. Methodology: This study included 14 cases diagnosed with OSCC. Following resection, the primary lesion was subjected to Histopathological analysis. 2 sets of HP slides were prepared from serial sectioning of the wax block prepared for each of the four margins. Both sets of slides were stained with H &E stain. One set of these slides was further stained with Pan CK marker (IHC) which is a cytokeratin marker to identify tumour cells. Results: All the slides with H & E staining reported negative for tumor infiltration and 4 slides (3 patients) out of 56 were reported positive with PanCK marker. There was a statistically significant difference in the number of patients with positive margins using IHC as compared to H & E stain. Conclusion: Immunohistochemistry using PanCK marker proved to be more efficient in the determination of status of safe margins than routine H & E staining.

ResSAT: Enhancing Spatial Transcriptomics Prediction from H&E- Stained Histology Images with Interactive Spot Transformer.

Spatial transcriptomics (ST) revolutionizes RNA quantification with high spatial resolution. Hematoxylin and eosin (H&E) images, the gold standard in medical diagnosis, offer insights into tissue structure, correlating with gene expression patterns. Current methods for predicting spatial gene expression from H&E images often overlook spatial relationships. We introduce ResSAT (Residual networks - Self-Attention Transformer), a framework generating spatially resolved gene expression profiles from H&E images by capturing tissue structures and using a self-attention transformer to enhance prediction.Benchmarking on 10× Visium datasets, ResSAT significantly outperformed existing methods, promising reduced ST profiling costs and rapid acquisition of numerous profiles.

Training immunophenotyping deep learning models with the same-section ground truth cell label derivation method improves virtual staining accuracy.

Introduction: Deep learning (DL) models predicting biomarker expression in images of hematoxylin and eosin (H&E)-stained tissues can improve access to multi-marker immunophenotyping, crucial for therapeutic monitoring, biomarker discovery, and personalized treatment development. Conventionally, these models are trained on ground truth cell labels derived from IHC-stained tissue sections adjacent to H&E-stained ones, which might be less accurate than labels from the same section. Although many such DL models have been developed, the impact of ground truth cell label derivation methods on their performance has not been studied. Methodology: In this study, we assess the impact of cell label derivation on H&E model performance, with CD3+ T-cells in lung cancer tissues as a proof-of-concept. We compare two Pix2Pix generative adversarial network (P2P-GAN)-based virtual staining models: one trained with cell labels obtained from the same tissue section as the H&E-stained section (the 'same-section' model) and one trained on cell labels from an adjacent tissue section (the 'serial-section' model). Results: We show that the same-section model exhibited significantly improved prediction performance compared to the 'serial-section' model. Furthermore, the same-section model outperformed the serial-section model in stratifying lung cancer patients within a public lung cancer cohort based on survival outcomes, demonstrating its potential clinical utility. Discussion: Collectively, our findings suggest that employing ground truth cell labels obtained through the same-section approach boosts immunophenotyping DL solutions.

Nanopore-based targeted next-generation sequencing of tissue samples for tuberculosis diagnosis.

Objective: Diagnosing tuberculosis (TB) can be particularly challenging in the absence of sputum for pulmonary tuberculosis cases and extrapulmonary TB (EPTB). This study evaluated the utility of nanopore-based targeted next-generation sequencing (tNGS) for diagnosing TB in tissue samples, and compared its efficacy with other established diagnostic methods. Methods: A total of 110 tissue samples from clinical cases were examined. The sensitivity and specificity of tNGS were benchmarked against a range of existing diagnostic approaches including hematoxylin and eosin (HE) staining in conjunction with acid-fast bacilli (AFB) detection, HE staining combined with PCR, HE staining paired with immunohistochemistry (IHC) using anti-MPT64, and the Xpert Mycobacterium tuberculosis (MTB)/rifampicin (RIF) assay. Results: The sensitivity and specificity of tNGS were 88.2 and 94.1%, respectively. The respective sensitivities for HE staining combined with AFB, HE staining combined with PCR, HE staining combined with IHC using anti-MPT64, and Xpert MTB/RIF were 30.1, 49.5, 47.3, and 59.1%. The specificities for these methods were 82.4, 88.2, 94.1, and 94.1%, respectively. Analysis of drug resistance based on tNGS results indicated that 10 of 93 TB patients (10.75%) had potential drug resistance. Conclusion: Targeted next-generation sequencing achieved higher accuracy than other established diagnostic methods, and can play a crucial role in the rapid and accurate diagnosis of TB, including drug-resistant TB.

Novel Artificial Intelligence Combining Convolutional Neural Network and Support Vector Machine to Predict Colorectal Cancer Prognosis and Mutational Signatures From Hematoxylin and Eosin Images.

Reducing recurrence following radical resection of colon cancer without overtreatment or undertreatment remains a challenge. Postoperative adjuvant chemotherapy (Adj) is currently administered based solely on pathologic TNM stage. However, prognosis can vary significantly among patients with the same disease stage. Therefore, novel classification systems in addition to the TNM are necessary to inform decision-making regarding postoperative treatment strategies, especially stage II and III disease, and minimize overtreatment and undertreatment with Adj. We developed a prognostic prediction system for colorectal cancer using a combined convolutional neural network and support vector machine approach to extract features from hematoxylin and eosin staining images. We combined the TNM and our artificial intelligence (AI)-based classification system into a modified TNM-AI classification system with high discriminative power for recurrence-free survival. Furthermore, the cancer cell population recognized by this system as low risk of recurrence exhibited the mutational signature SBS87 as a genetic phenotype. The novel AI-based classification system developed here is expected to play an important role in prognostic prediction and personalized treatment selection in oncology.

Establishment of a rabbit model of different degrees of testicular torsion.

INTRODUCTION AND OBJECTIVES: Different degrees of testicular torsion result in varying degrees of testicular damage, which influences treatment options and outcomes. Therefore, establishing a testicular torsion model with different degrees is necessary for clinical diagnosis. MATERIALS AND METHODS: Rabbits were randomly divided into four groups and their spermatic cords were twisted at 0 °, 180 °, 360 °, and 720 °, respectively. Color Doppler flow imaging (CDFI) were performed to evaluate the blood supply in testicles. The twisted testicles were surgically removed at six hours post-operation and were evaluated by morphological observation and Hematoxylin and Eosin staining. RESULTS: CDFI signals were gradually decreased as the degree of testicular torsion increased, and scores of CDFI in the 360 ° and 720 ° groups were significantly decreased at postoperative six hours compared to pre-surgery. Compared to the sham, the testicle in the 180 ° group exhibited slight congestion, whereas the testicles in the 360 ° and 720 ° groups were dark red in color and had severe congestion and unrecognizable vessels. Hematoxylin and Eosin staining showed mild spermatogenic cell reduction and testicular interstitial hemorrhage in the 180 ° group. In the 360 ° and 720 ° groups, disordered seminiferous tubules, shed spermatogenic cells in tubules, inflammatory cell infiltration, and severe hemorrhage were found. In comparison with the sham, interstitial hemorrhage scores in the 360 ° and 720 ° groups were significantly higher, and scores of germinal epithelial cell thickness in the three testicular torsion groups were significantly decreased. CONCLUSIONS: Collectively, we successfully constructed a testicular torsion model with different degrees in rabbits.

A simple and economical method for unbiasedly quantifying the alveolar size of entire mouse lung tissue utilizing Hematoxylin and Eosin Staining.

Alveolar, the smallest structural and functional units within the respiratory system, play a crucial role in maintaining lung function. Alveolar damage is a typical pathological hallmark of respiratory diseases. Nevertheless, there is currently no simple, rapid, economical, and unbiased method for quantifying alveolar size for entire lung tissue. Here, firstly, we conducted lung sample slicing based on the size, shape, and distribution of airway branches of different lobes. Next, we performed HE staining on different slices. Then, we provided an unbiased quantification of alveolar size using free software ImageJ. Through this protocol, we demonstrated that C57Bl/6 mice exhibit varying alveolar sizes among different lobes. Collectively, we provided a simple and unbiased method for a more comprehensive quantification of alveolar size in mice, which holds promise for a broader range of respiratory research using mouse models.

Dapagliflozin mitigates oxidative stress, inflammatory, and histopathological markers of aging in mice.

Aging, a complex physiological process affecting all living things, is a major area of research, particularly focused on interventions to slow its progression. This study assessed the antiaging efficacy of dapagliflozin (DAPA) on various aging-related parameters in a mouse model artificially induced to age. Forty male Swiss albino mice were randomly divided into four groups of ten animals each. The control group (Group I) received normal saline. The aging model group (Group II) was administered D-galactose orally at 500mg/kg to induce aging. Following the aging induction, the positive control group received Vitamin C supplementation (Group III), while the DAPA group (Group IV) was treated with dapagliflozin. The inflammatory mediators (TNF-α and IL-1ß) showed similar patterns of change. No statistically significant difference was observed between groups III and IV. Both groups had significantly lower values compared to GII, while it was significantly higher compared to GI. Glutathione peroxidase (GSH-Px) showed no statistically significant difference between groups GIII and GIV, but it was higher in GIII compared to GII and significantly lower in GIII compared to GI. The study demonstrated that dapagliflozin exerts a beneficial impact on many indicators of aging in mice. The intervention resulted in a reduction in hypertrophy in cardiomyocytes, an enhancement in skin vitality, a decrease in the presence of inflammatory mediators, and an improvement in the efficacy of antioxidants.

Prognostic significance of Klotho expression in patients with renal cell carcinoma.

INTRODUCTION: Various molecular markers have been investigated in renal cell carcinoma (RCC) without significant reliability. We analyzed Klotho (tumor suppressive protein) expression in RCC to investigate its association with tumor-stage, grade, disease-free-survival (DFS) and overall-survival (OS). METHODS: Data of histologically confirmed patients of RCC with complete clinical follow-up were retrieved from Medical-Record-Library. Tissue sections of tumor and normal parenchyma were prepared from the blocks. Immunohistochemical studies for Klotho were done with commercially available kit (EPR6856, Ab181373; Abcam, Cambridge MA, USA). Klotho expression was scored between 0-3 and grouped into weak/absent (0, 1) and moderate/strong (2, 3). Tumors stages and grades were grouped into low stage (I and II) and high stage (III and IV) and into low grade (grade 1 and 2) and high grade (grade 3 and 4) according to WHO/ISUP grading. The histopathologists were blinded as to the clinical and follow-up data. Various prognostic factors were analyzed with respect to Klotho expression. Kaplan-Meier curves were created for DFS and OS. RESULTS: Fifty-four patients of mean age 55.15 ± 13.34 years and M:F ratio of 1.8:1 were included. Normal renal tissue had strong expression of Klotho in all. In tumor tissue 20 (37%) had negative, 7 (13%) had weak, 14 (25.9%) had moderate and 13 (24.1%) had strong Klotho expression. Significantly more patients had absent/weak Klotho expression with higher grade (16/24 (66.7%) vs 7/25 (28%); p = 0.007), higher stage (22/33 (66%) vs 5/21 (23.8%); p = 0.002), LVI (12/14 (85.7%) vs 2/14 (14.3%); p = 0.002), sinus-fat-invasion (16/21 (76.2%) vs 5/21 (23.8%); p = 0.002), renal-vein-involvement (14/18 (77.8%) vs 4/18 (22.2%); p = 0.004), necrosis (17/26 (65.3%) vs 9/26 (34.6%); p = 0.029) and metastasis (8/9 (88.9%) vs 1/9 (11.1%); p = 0.01). Median DFS and OS were significantly lower in patients with weak/absent Klotho expression (12 vs 23 months, p = 0.023 and 15 vs 33 months, p = 0.006 respectively). Kaplan-Meier curves showed lower estimated DFS and OS in patients with weak/absent expression. CONCLUSIONS: We conclude that Klotho expression in renal tumor could be a good prognostic marker in patients with RCC.

Pediatric upper lip myopericytoma: a case report and comprehensive review.

BACKGROUND: Myopericytoma is a rare spindle cell tumor of mesenchymal origin, typically benign, characterized by concentric proliferation of tumor cells around blood vessels within subcutaneous tissue. It primarily occurs in middle-aged adults and is often located in distal extremities, although cases have been reported in proximal extremities and head-neck regions. However, occurrences within the oral cavity are exceedingly rare. To date, literature reviews have identified only two cases in children under 10 years old and reported only five cases of myopericytoma occurring in the lip region. We provide a comprehensive review and analysis of all documented cases to better understand this condition. CASE PRESENTATION: A 7-year-old girl presented to oral and maxillofacial surgery with the discovery of a painless mass on the inner aspect of the upper lip. The diagnosis of myopericytoma was confirmed by histological examination (HE staining), alcian blue staining, and immunohistochemistry. CONCLUSIONS: Following surgical excision, there were no signs of recurrence at a 3-month follow-up. The pathological diagnosis of myopericytoma is quite challenging, and immunohistochemical testing is necessary.

Hepatic and renal effects of oral stingless bee honey in a streptozotocin-induced diabetic rat model.

BACKGROUND: Diabetes is known damage the liver and kidney, leading to hepatic dysfunction and kidney failure. Honey is believed to help in lowering the blood glucose levels of diabetic patients and reducing diabetic complications. However, the effect of stingless bee honey (SBH) administration in relieving liver and kidney damage in diabetes has not been well-studied. AIM: To investigate the effect of SBH administration on the kidney and liver of streptozotocin-induced (STZ; 55 mg/kg) diabetic Sprague Dawley rats. METHODS: The rats were grouped as follows (n = 6 per group): non-diabetic (ND), untreated diabetic (UNT), metformin-treated (MET), and SBH+metformin-treated (SBME) groups. After successful diabetic induction, ND and UNT rats were given normal saline, whereas the treatment groups received SBH (2.0 g/kg and/or metformin (250 mg/kg) for 12 d. Serum biochemical parameters and histological changes using hematoxylin and eosin (H&E) and periodic acid-Schiff (PAS) staining were evaluated. RESULTS: On H&E and PAS staining, the ND group showed normal architecture and cellularity of Bowman's capsule and tubules, whereas the UNT and MET groups had an increased glomerular cellularity and thickened basement membrane. The SBH-treated group showed a decrease in hydropic changes and mild cellularity of the glomerulus vs the ND group based on H&E staining, but the two were similar on PAS staining. Likewise, the SBME-treated group had an increase in cellularity of the glomerulus on H&E staining, but it was comparable to the SBH and ND groups on PAS staining. UNT diabetic rats had tubular hydropic tubules, which were smaller than other groups. Reduced fatty vacuole formation and dilated blood sinusoids in liver tissue were seen in the SBH group. Conversely, the UNT group had high glucose levels, which subsequently increased MDA levels, ultimately leading to liver damage. SBH treatment reduced this damage, as evidenced by having the lowest fasting glucose, serum alanine transaminase, aspartate transaminase, and alkaline phosphatase levels compared to other groups, although the levels of liver enzymes were not statistically significant. CONCLUSION: The cellularity of the Bowman's capsule, as well as histological alteration of kidney tubules, glomerular membranes, and liver tissues in diabetic rats after oral SBH resembled those of ND rats. Therefore, SBH exhibited a protective hepatorenal effect in a diabetic rat model.

Transparency-enhancing technology allows the three-dimensional assessment of esophageal carcinoma obtained by endoscopic submucosal dissection.

BACKGROUND: Although much progress has been made in diagnosis of carcinomas, no established methods have been confirmed to elucidate their morphological features. METHODS: Three-dimensional structure of esophageal carcinomas was assessed using transparency-enhancing technology. Endoscopically resected esophageal squamous cell carcinoma was fluorescently stained, optically cleared using a transparency-enhancing reagent called LUCID, and visualized using laser scanning microscopy. The resulting microscope images were converted to virtual HE images for observation using ImageJ software. RESULTS: Microscopic observation and image editing enabled three-dimensional image reconstruction and conversion to virtual HE images. The structure of abnormal blood vessels in esophageal carcinoma recognized by endoscopy could be observed in the 3 dimensions. Squamous cell carcinoma and normal squamous epithelium could be distinguished in the virtual HE images. CONCLUSIONS: The results suggested that transparency-enhancing technology and virtual HE images may be feasible for clinical application and represent a novel histopathological method for evaluating endoscopically resected specimens.

The Mechanism of Astragaloside IV in NOD-like Receptor Family Pyrin Domain Containing 3 Inflammasome-mediated Pyroptosis after Intracerebral Hemorrhage.

BACKGROUND: Intracerebral hemorrhage (ICH) is one of the most common subtypes of stroke. OBJECTIVES: This study aimed to investigate the mechanism of Astragaloside IV (AS-IV) on inflammatory injury after ICH. METHODS: The ICH model was established by the injection of collagenase and treated with ASIV (20 mg/kg or 40 mg/kg). The neurological function, water content of the bilateral cerebral hemisphere and cerebellum, and pathological changes in brain tissue were assessed. The levels of interleukin-1 beta (IL-1ß), IL-18, tumor necrosis factor-alpha, interferon-gamma, and IL-10 were detected by enzyme-linked immunosorbent assay. The levels of Kruppel-like factor 2 (KLF2), NOD-like receptor family pyrin domain containing 3 (NLRP3), GSDMD-N, and cleaved-caspase-1 were detected by reverse transcription-quantitative polymerase chain reaction and Western blot assay. The binding relationship between KLF2 and NLRP3 was verified by chromatin-immunoprecipitation and dual-luciferase assays. KLF2 inhibition or NLRP3 overexpression was achieved in mice to observe pathological changes. RESULTS: The decreased neurological function, increased water content, severe pathological damage, and inflammatory response were observed in mice after ICH, with increased levels of NLRP3/GSDMD-N/cleaved-caspase-1/IL-1ß/IL-18 and poorly-expressed KLF2 in brain tissue. After AS-IV treatment, the neurological dysfunction, high brain water content, inflammatory response, and pyroptosis were alleviated, while KLF2 expression was increased. KLF2 bonded to the NLRP3 promoter region and inhibited its transcription. Down-regulation of KLF2 or upregulation of NLRP3 reversed the effect of AS-IV on inhibiting pyroptosis and reducing inflammatory injury in mice after ICH. CONCLUSION: AS-IV inhibited NLRP3-mediated pyroptosis by promoting KLF2 expression and alleviated inflammatory injury in mice after ICH.

Morphology-based molecular classification of spinal cord ependymomas using deep neural networks.

Based on DNA-methylation, ependymomas growing in the spinal cord comprise two major molecular types termed spinal (SP-EPN) and myxopapillary ependymomas (MPE(-A/B)), which differ with respect to their clinical features and prognosis. Due to the existing discrepancy between histomorphogical diagnoses and classification using methylation data, we asked whether deep neural networks can predict the DNA methylation class of spinal cord ependymomas from hematoxylin and eosin stained whole-slide images. Using explainable AI, we further aimed to prospectively improve the consistency of histology-based diagnoses with DNA methylation profiling by identifying and quantifying distinct morphological patterns of these molecular ependymoma types. We assembled a case series of 139 molecularly characterized spinal cord ependymomas (nMPE = 84, nSP-EPN = 55). Self-supervised and weakly-supervised neural networks were used for classification. We employed attention analysis and supervised machine-learning methods for the discovery and quantification of morphological features and their correlation to the diagnoses of experienced neuropathologists. Our best performing model predicted the DNA methylation class with 98% test accuracy and used self-supervised learning to outperform pretrained encoder-networks (86% test accuracy). In contrast, the diagnoses of neuropathologists matched the DNA methylation class in only 83% of cases. Domain-adaptation techniques improved model generalization to an external validation cohort by up to 22%. Statistically significant morphological features were identified per molecular type and quantitatively correlated to human diagnoses. The approach was extended to recently defined subtypes of myxopapillary ependymomas (MPE-(A/B), 80% test accuracy). In summary, we demonstrated the accurate prediction of the DNA methylation class of spinal cord ependymomas (SP-EPN, MPE(-A/B)) using hematoxylin and eosin stained whole-slide images. Our approach may prospectively serve as a supplementary resource for integrated diagnostics and may even help to establish a standardized, high-quality level of histology-based diagnostics across institutions-in particular in low-income countries, where expensive DNA-methylation analyses may not be readily available.

HoLy-Net: Segmentation of histological images of diffuse large B-cell lymphoma.

Over the last years, there has been large progress in automated segmentation and classification methods in histological whole slide images (WSIs) stained with hematoxylin and eosin (H&E). Current state-of-the-art (SOTA) techniques are based on diverse datasets of H&E-stained WSIs of different types of predominantly solid cancer. However, there is a scarcity of methods and datasets enabling segmentation of tumors of the lymphatic system (lymphomas). Here, we propose a solution for segmentation of diffuse large B-cell lymphoma (DLBCL), the most common non-Hodgkin's lymphoma. Our method applies to both H&E-stained slides and to a broad range of markers stained with immunohistochemistry (IHC). While IHC staining is an important tool in cancer diagnosis and treatment decisions, there are few automated segmentation and classification methods for IHC-stained WSIs. To address the challenges of nuclei segmentation in H&E- and IHC-stained DLBCL images, we propose HoLy-Net - a HoVer-Net-based deep learning model for lymphoma segmentation. We train two different models, one for segmenting H&E- and one for IHC-stained images and compare the test results with the SOTA methods as well as with the original version of HoVer-Net. Subsequently, we segment patient WSIs and perform single cell-level analysis of different cell types to identify patient-specific tumor characteristics such as high level of immune infiltration. Our method outperforms general-purpose segmentation methods for H&E staining in lymphoma WSIs (with an F1 score of 0.899) and is also a unique automated method for IHC slide segmentation (with an F1 score of 0.913). With our solution, we provide a new dataset we denote LyNSeC (lymphoma nuclear segmentation and classification) containing 73,931 annotated cell nuclei from H&E and 87,316 from IHC slides. Our method and dataset open up new avenues for quantitative, large-scale studies of morphology and microenvironment of lymphomas overlooked by the current automated segmentation methods.

An orientation-free ring feature descriptor with stain-variability normalization for pathology image matching.

Comprehensively analyzing the corresponding regions in the images of serial slices stained using different methods is a common but important operation in pathological diagnosis. To help increase the efficiency of the analysis, various image registration methods are proposed to match the corresponding regions in different images, but their performance is highly influenced by the rotations, deformations, and variations of staining between the serial pathology images. In this work, we propose an orientation-free ring feature descriptor with stain-variability normalization for pathology image matching. Specifically, we normalize image staining to similar levels to minimize the impact of staining differences on pathology image matching. To overcome the rotation and deformation issues, we propose a rotation-invariance orientation-free ring feature descriptor that generates novel adaptive bins from ring features to build feature vectors. We measure the Euclidean distance of the feature vectors to evaluate keypoint similarity to achieve pathology image matching. A total of 46 pairs of clinical pathology images in hematoxylin-eosin and immunohistochemistry straining to verify the performance of our method. Experimental results indicate that our method meets the pathology image matching accuracy requirements (error ¡ 300µm), especially competent for large-angle rotation cases common in clinical practice.

MGGAN: A multi-generator generative adversarial network for breast cancer immunohistochemical image generation.

The immunohistochemical technique (IHC) is widely used for evaluating diagnostic markers, but it can be expensive to obtain IHC-stained section. Translating the cheap and easily available hematoxylin and eosin (HE) images into IHC images provides a solution to this challenge. In this paper, we propose a multi-generator generative adversarial network (MGGAN) that can generate high-quality IHC images based on the HE of breast cancer. Our MGGAN approach combines the low-frequency and high-frequency components of the HE image to improve the translation of breast cancer image details. We use the multi-generator to extract semantic information and a U-shaped architecture and patch-based discriminator to collect and optimize the low-frequency and high-frequency components of an image. We also include a cross-entropy loss as a regularization term in the loss function to ensure consistency between the synthesized image and the real image. Our experimental and visualization results demonstrate that our method outperforms other state-of-the-art image synthesis methods in terms of both quantitative and qualitative analysis. Our approach provides a cost-effective and efficient solution for obtaining high-quality IHC images.

Study of radiochemotherapy decision-making for young high-risk low-grade glioma patients using a macroscopic and microscopic combined radiomics model.

OBJECTIVES: As a few types of glioma, young high-risk low-grade gliomas (HRLGGs) have higher requirements for postoperative quality of life. Although adjuvant chemotherapy with delayed radiotherapy is the first treatment strategy for HRLGGs, not all HRLGGs benefit from it. Accurate assessment of chemosensitivity in HRLGGs is vital for making treatment choices. This study developed a multimodal fusion radiomics (MFR) model to support radiochemotherapy decision-making for HRLGGs. METHODS: A MFR model combining macroscopic MRI and microscopic pathological images was proposed. Multiscale features including macroscopic tumor structure and microscopic histological layer and nuclear information were grabbed by unique paradigm, respectively. Then, these features were adaptively incorporated into the MFR model through attention mechanism to predict the chemosensitivity of temozolomide (TMZ) by means of objective response rate and progression free survival (PFS). RESULTS: Macroscopic tumor texture complexity and microscopic nuclear size showed significant statistical differences (p < 0.001) between sensitivity and insensitivity groups. The MFR model achieved stable prediction results, with an area under the curve of 0.950 (95% CI: 0.942-0.958), sensitivity of 0.833 (95% CI: 0.780-0.848), specificity of 0.929 (95% CI: 0.914-0.936), positive predictive value of 0.833 (95% CI: 0.811-0.860), and negative predictive value of 0.929 (95% CI: 0.914-0.934). The predictive efficacy of MFR was significantly higher than that of the reported molecular markers (p < 0.001). MFR was also demonstrated to be a predictor of PFS. CONCLUSIONS: A MFR model including radiomics and pathological features predicts accurately the response postoperative TMZ treatment. CLINICAL RELEVANCE STATEMENT: Our MFR model could identify young high-risk low-grade glioma patients who can have the most benefit from postoperative upfront temozolomide (TMZ) treatment. KEY POINTS: • Multimodal radiomics is proposed to support the radiochemotherapy of glioma. • Some macro and micro image markers related to tumor chemotherapy sensitivity are revealed. • The proposed model surpasses reported molecular markers, with a promising area under the curve (AUC) of 0.95.

AI Deployment on GBM Diagnosis: A Novel Approach to Analyze Histopathological Images Using Image Feature-Based Analysis.

BACKGROUND: Glioblastoma (GBM) is one of the most common malignant primary brain tumors, which accounts for 60-70% of all gliomas. Conventional diagnosis and the decision of post-operation treatment plan for glioblastoma is mainly based on the feature-based qualitative analysis of hematoxylin and eosin-stained (H&E) histopathological slides by both an experienced medical technologist and a pathologist. The recent development of digital whole slide scanners makes AI-based histopathological image analysis feasible and helps to diagnose cancer by accurately counting cell types and/or quantitative analysis. However, the technology available for digital slide image analysis is still very limited. This study aimed to build an image feature-based computer model using histopathology whole slide images to differentiate patients with glioblastoma (GBM) from healthy control (HC). METHOD: Two independent cohorts of patients were used. The first cohort was composed of 262 GBM patients of the Cancer Genome Atlas Glioblastoma Multiform Collection (TCGA-GBM) dataset from the cancer imaging archive (TCIA) database. The second cohort was composed of 60 GBM patients collected from a local hospital. Also, a group of 60 participants with no known brain disease were collected. All the H&E slides were collected. Thirty-three image features (22 GLCM and 11 GLRLM) were retrieved from the tumor volume delineated by medical technologist on H&E slides. Five machine-learning algorithms including decision-tree (DT), extreme-boost (EB), support vector machine (SVM), random forest (RF), and linear model (LM) were used to build five models using the image features extracted from the first cohort of patients. Models built were deployed using the selected key image features for GBM diagnosis from the second cohort (local patients) as model testing, to identify and verify key image features for GBM diagnosis. RESULTS: All five machine learning algorithms demonstrated excellent performance in GBM diagnosis and achieved an overall accuracy of 100% in the training and validation stage. A total of 12 GLCM and 3 GLRLM image features were identified and they showed a significant difference between the normal and the GBM image. However, only the SVM model maintained its excellent performance in the deployment of the models using the independent local cohort, with an accuracy of 93.5%, sensitivity of 86.95%, and specificity of 99.73%. CONCLUSION: In this study, we have identified 12 GLCM and 3 GLRLM image features which can aid the GBM diagnosis. Among the five models built, the SVM model proposed in this study demonstrated excellent accuracy with very good sensitivity and specificity. It could potentially be used for GBM diagnosis and future clinical application.