Prediction of clinically actionable genetic alterations from colorectal cancer histopathology images using deep learning

BACKGROUND Identifying genetic mutations in cancer patients have been increasingly important because distinctive mutational patterns can be very informative to determine the optimal therapeutic strategy. Recent studies have shown that deep learning-based molecular cancer subtyping can be performed directly from the standard hematoxylin and eosin(H&E) sections in diverse tumors including colorectal cancers(CRCs). Since H&E-stained tissue slides are ubiquitously available, mutation prediction with the pathology images from cancers can be a time-and cost-effective complementary method for personalized treatment.AIM To predict the frequently occurring actionable mutations from the H&E-stained CRC whole-slide images(WSIs) with deep learning-based classifiers.METHODS A total of 629 CRC patients from The Cancer Genome Atlas(TCGA-COAD and TCGA-READ) and 142 CRC patients from Seoul St. Mary Hospital(SMH) were included. Based on the mutation frequency in TCGA and SMH datasets, we chose APC, KRAS, PIK3CA, SMAD4, and TP53 genes for the study. The classifiers were trained with 360 × 360 pixel patches of tissue images. The receiver operating characteristic(ROC) curves and area under the curves(AUCs) for all the classifiers were presented.RESULTS The AUCs for ROC curves ranged from 0.693 to 0.809 for the TCGA frozen WSIs and from 0.645 to 0.783 for the TCGA formalin-fixed paraffin-embedded WSIs.The prediction performance can be enhanced with the expansion of datasets. When the classifiers were trained with both TCGA and SMH data, the prediction performance was improved.CONCLUSION APC, KRAS, PIK3CA, SMAD4, and TP53 mutations can be predicted from H&E pathology images using deep learning-based classifiers, demonstrating the potential for deep learning-based mutation prediction in the CRC tissue slides.

The Comparison of the Manifestation of the Clinical Imageology and Pathology between the Brucellar Spondylitis and the Spine Turberculosis

Objective: To improve the clinical differential diagnosis level, the clinical manifestation of the brucellar spondylitis and the spine turberculosis were discussed in this paper. Method: The study was completed in the No. 1 Affiliated Hospital of Hebei North University in Zhangjiakou City, Hebei Province, China, from January 2001 to December 2013 by Analyzing the X-ray, CT scanning and MRI of 257 cases of the brucellar spondylitis retrospectively and comparing with the clinical imageology and pathology 332 cases of turberculosis of the spine diagnosed finally. Results: The brucellar spondylitis: The focuses usually locate in the lumbar vertebra and L4, 5 has the highest occurrence rate. The focuses are often small but multiple, and limited to the edge of the vertebra. Hyperostosis and osteoscterosis are usually found in the tissuses around the focuses. There are often new focuses in the newborn bones, and the destruction of intervertebral discs is usually slight. Hyperostosis and osteoscterosis might be found in the surfaces of the joints. The densites of the bones close to the focuses become high. There were less or no paravertebral abscesses but inflammational granuloma can be found frequently. Turberculosis of the spine: The focuses are usually located in the thoracic and lumbar vertebra, and are characterized by the destruction of the vertebra and the intervertebral discs, accompanied by the appearance of dead bones. In most cases, paravertebral abscesses and osteoporosis might be found. Conclusions: The specific manifestation of the clinical imageology can help to differentiate the brucelar spondylitis from the turberculosis of the spine.

LUPUS NEPHRITIS COMPLICATED WITH MALIGNANT HYPERTENSION:FROM RENAL VASCULAR PATHOLOGY TO CLINICAL RELEVANCE

Objective To investigate the clinical and pathological characteristics of lupus nephritis patients complicated with malignant hypertension.Methods We retrospectively studied 19 patients with lupus nephritis complicated with malignant hypertension who underwent renal biopsy between January 2002 and December 2006.Results Of 19 patients,3 were men and 16 were women,with a mean age of 24.4±7.7 years old.All had positive antinuclear antibodies and low serum complement was found in 13 patients.All were anemic and 12 of them were thrombocytopenic.Impaired renal function was found in 17 patients with an average serum creatinine of 184.5±88.9 μmol/L.Severe intrarenal arteriolar lesion was found in all patients.Six patients had lupus vasculopathy,11 patients had renal thrombotic microangiopathy lesion,2 had severe arteriosclerosis.All patients received steroids and immunosuppressive drugs,15 received angiotensin-converting enzyme inhibitor(ACEI)/angiotensin receptor blocker(ARB)with resultant well-controlled blood pressure.Thrombocytopenia and hemolytic anemia resolved remarkably.The renal function improved or recovered in 14 of 17 patients,and 3 developed end-stage renal disease on maintenance dialysis.Conclusions Severe intrarenal vascular lesion complicated with renal nephritis parallels clinical manifestation of malignant hypertension.Renal pathology is the key of treatment strategy emphasizing on the significance of renal vascular involvement and type.On the basis of immunosuppressive drugs and steroids to control systemic lupus activity,timely initiation of ACEI/ARB could be of benefit to blood pressure control and long term renal survival.

Developing a better mouse model of Alzheimer disease with clinically relevant phenotypes in tau pathology

OBJECTIVE Transgenic mouse model has been widely used in pathogenesis study and preclinical drug evaluation in Alzheimer disease(AD).However,key differences are found between current animal models and clinical AD patients regarding phenotypes.Lack of complete models that recapitulate broad spectrum of human AD neuropathology restricts efficacy of research projects and leads to frequent failure in AD drug development at clinical trial stages.This study aims to develop better mouse models of AD through modifying key phenotype insufficiency.METHODS By crossing different single and double transgenic mice with different mutations of APP/PS1 or tau and under prion,Thy1 or PDGF-β promoter,as well as selected knockout mice,I produced a dozen of bigenic models for neuropathology screening.Further neurochemical,behavioral and pharmacological validations were conducted in the optimized mouse model.RESULTS Neuropathology phenotyping found remarkable differences in tau pathology and neurodegeneration among individual APP/PS1/tau transgenic models.I had identified a triple mouse model named FADT that showed(1) huge mature tau pathology in hippocampus and cortex;(2) abundant tau truncation,as seen in human AD brain;(3)progressive neurodegeneration;(4)selective brain atrophy in hippocampus and entorhinal cortex;(5) reproducible and late onset spatial memory defects,etc.Importantly,remarkable tau pathology in this FADT model is mainly driven by beta-amyloid pathology,which differs from high expression of tau in rTg4510 model.CONCLUSION I had developed a new triple transgenic mouse model that recapitulates broad spectrum of human AD neuropathology features.This study will not only establish a solid model basis for AD pathophysiology investigation and drug development,but also reveal important clues on the interaction of beta-amyloid and tau pathologies in the brain.

Clinical pathology analysis of esophageal sarcomatoid carcinoma

Objective： The purpose of this study was to study the clinical, imaging characters and pathological characteristics of esophageal sarcomatoid carcinoma. Methods： We reviewed 23 cases of esophageal sarcomatoid carcinoma from Janu ary 2006 to December 2013 in four hospitals. The data of patients who were esophageal sarcomatoid carcinoma operated were retrospectively analyzed. All cases had completed upper gastrointestinal barium images materials and 14 of these cases had completed CT images materials. Upper gastrointestinal barium images and CT imaging features include tumor location, size, shape, and strengthen, etc. The biological parameters of lesions including the express of cytokeratin AE1/AE3, 34E12,, Vimentin, desmin, Actin, S100 and Ki67 detected by immunhistochemical UltraSensitiveTM SP method （n = 23）, and the patients＇ data of contrastographic picture （n = 23）, imaging characters of CT scan （n = 14）, and their relationship were studied. Results： Upper gastrointestinal barium images, CT imaging and gastrointestinal fiberscopy revealed Iobulated intraluminal filling defect 0.4 cm to 5.7 cm x 3.5 cmx 1.3 cm （mean = 3. 7 cm） in the mid （n = 14）, lower （n = 7） and upper （n = 2） intrathoracic esophagus. Among 23 cases of esophageal sarcomatoid carcinoma, 19 patients were of mushroom type, 2 patients was of ulcer type, and 2 patients were of medulla type; 19 patients were pedunculated, and 4 patients were no pedunculated （2 patients was of ulcer type）. The tumor surface was relatively smooth and esophageal compliance was maintained. The pathological changes of esophagus such as lightly locked, rigid wall nomanifest partly, esophageal lumens expand partly, major filling sublobe defect could be shown through contrast medium. Normal esophagus was no unpack obviously over pathological changes. Enhanced computed tomography showed tumors in the intrathoracic esophagus and 8 lymph nodes metastases in 3 cases. Histologically, carcinomatous and sarcomatous components coexist. Microscopically, the tumor comprised poorly differentiated squamous cell carcinoma and spindleshaped cells resembling leiomyosarcoma. Immu nohistochemically, spindleshaped sarcomatous cells displayed weekly positive reaction to cytokeratin AE1/AE3. Transitional zone was seen between sarcomatous and carcinomatous elements in 5 cases. The 17 lymph nodes metastases in 5 cases （53 lymph nodes） among 23 cases esophageal sarcomatoid carcinoma （187 lymph nodes） were observed. Conclusion： The clinical and radiologic features of esophageal sarcomatoid carcinoma overlap with those of other esophageal neoplasms. There are the radiologic imaging changes such as a large, intraluminal, polypoid mass, major filling sublobe defect and pedicle skin flap tumor in esophageal lumen, esophageal lumen extension partly, dissepiment rigidity wall no obviously, etc. Histologically, carcinomatous and sarcomatous components coexist and the biphasic pattern is the key diagnostic feature. However, esophageal sarcomatoid carcinoma has a more favorable prognosis than other malignant esophageal neoplasms. Immunohistochemical staining seems necessary to distinguish these lesions from other esophageal neoplasms.

What benefit can be obtained from magnetic resonance imaging diagnosis with artificial intelligence in prostate cancer compared with clinical assessments?

The present study aimed to explore the potential of artificial intelligence(AI)methodology based on magnetic resonance(MR)images to aid in the management of prostate cancer(PCa).To this end,we reviewed and summarized the studies comparing the diagnostic and predictive performance for PCa between AI and common clinical assessment methods based on MR images and/or clinical characteristics,thereby investigating whether AI methods are generally superior to common clinical assessment methods for the diagnosis and prediction fields of PCa.First,we found that,in the included studies of the present study,AI methods were generally equal to or better than the clinical assessment methods for the risk assessment of PCa,such as risk stratification of prostate lesions and the prediction of therapeutic outcomes or PCa progression.In particular,for the diagnosis of clinically significant PCa,the AI methods achieved a higher summary receiver operator characteristic curve(SROC-AUC)than that of the clinical assessment methods(0.87 vs.0.82).For the prediction of adverse pathology,the AI methods also achieved a higher SROC-AUC than that of the clinical assessment methods(0.86 vs.0.75).Second,as revealed by the radiomics quality score(RQS),the studies included in the present study presented a relatively high total average RQS of 15.2(11.0–20.0).Further,the scores of the individual RQS elements implied that the AI models in these studies were constructed with relatively perfect and standard radiomics processes,but the exact generalizability and clinical practicality of the AI models should be further validated using higher levels of evidence,such as prospective studies and open-testing datasets.

Methylation status of c-fms oncogene in HCC and its relationship with clinical pathology

INTRODUCTIONThe mechanism that DNA hypomethylation leads toactivation of oncogene and occurrence of malignantneoplasm is being increasingly recognized byresearchers. Normal DNA methylation playsimportant role in stabilizing the phenotype of cell.DNA methylation status reduction and/or patternalteration are related to activation and abnormallyhigh expression of some oncogenes and cellularmalignancy[1-6]. c-fms oncogene encodes for colonystimulating factor 1 receptor (CSF-1R)[7], c-fms/CSF-1R was highly expressed in hepatocellularcarcinoma (HCC) tissue, but the mechanismremained obscure[8,9].

Expanding role and scope of artificial intelligence in the field of gastrointestinal pathology

Digital pathology(DP)and its subsidiaries including artificial intelligence(AI)are rapidly making inroads into the area of diagnostic anatomic pathology(AP)including gastrointestinal(GI)pathology.It is poised to revolutionize the field of diagnostic AP.Historically,AP has been slow to adopt digital technology,but this is changing rapidly,with many centers worldwide transitioning to DP.Coupled with advanced techniques of AI such as deep learning and machine learning,DP is likely to transform histopathology from a subjective field to an objective,efficient,and transparent discipline.AI is increasingly integrated into GI pathology,offering numerous advancements and improvements in overall diagnostic accuracy,efficiency,and patient care.Specifically,AI in GI pathology enhances diagnostic accuracy,streamlines workflows,provides predictive insights,integrates multimodal data,supports research,and aids in education and training,ultimately improving patient care and outcomes.This review summarized the latest developments in the role and scope of AI in AP with a focus on GI pathology.The main aim was to provide updates and create awareness among the pathology community.

Clinical,muscle pathology and molecular genetic analysis of myofibrillar myopathy 3 associated with MYOT gene mutation

Objective To analyze the clinical phenotypic characteristics,muscle pathology,genetic mutations and related proteins of myofibrillar myopathy 3 caused by mutation in MYOT gene,and to conduct a literature review and summary of this disease.Methods A retrospective analysis of the clinical phenotypic characteristics.

Clinical manifestations,diagnosis and long-term prognosis of adult autoimmune enteropathy:Experience from Peking Union Medical College Hospital

BACKGROUND Autoimmune enteropathy(AIE)is a rare disease whose diagnosis and long-term prognosis remain challenging,especially for adult AIE patients.AIM To improve overall understanding of this disease’s diagnosis and prognosis.METHODS We retrospectively analyzed the clinical,endoscopic and histopathological characteristics and prognoses of 16 adult AIE patients in our tertiary medical center between 2011 and 2023,whose diagnosis was based on the 2007 diagnostic criteria.RESULTS Diarrhea in AIE patients was characterized by secretory diarrhea.The common endoscopic manifestations were edema,villous blunting and mucosal hyperemia in the duodenum and ileum.Villous blunting(100%),deep crypt lymphocytic infiltration(67%),apoptotic bodies(50%),and mild intraepithelial lymphocytosis(69%)were observed in the duodenal biopsies.Moreover,there were other remarkable abnormalities,including reduced or absent goblet cells(duodenum 94%,ileum 62%),reduced or absent Paneth cells(duodenum 94%,ileum 69%)and neutrophil infiltration(duodenum 100%,ileum 69%).Our patients also fulfilled the 2018 diagnostic criteria but did not match the 2022 diagnostic criteria due to undetectable anti-enterocyte antibodies.All patients received glucocorticoid therapy as the initial medication,of which 14/16 patients achieved a clinical response in 5(IQR:3-20)days.Immunosuppressants were administered to 9 patients with indications of steroid dependence(6/9),steroid refractory status(2/9),or intensified maintenance medication(1/9).During the median of 20.5 months of followup,2 patients died from multiple organ failure,and 1 was diagnosed with non-Hodgkin’s lymphoma.The cumulative relapse-free survival rates were 62.5%,55.6%and 37.0%at 6 months,12 months and 48 months,respectively.CONCLUSION Certain histopathological findings,including a decrease or disappearance of goblet and Paneth cells in intestinal biopsies,might be potential diagnostic criteria for adult AIE.The long-term prognosis is still unsatisfactory despite corticosteroid and immunosuppressant medications,which highlights the need for early diagnosis and novel medications.

Brain Pathology in COVID-19:Clinical Manifestations and Potential Mechanisms

Neurological manifestations of coronavirus disease 2019(COVID-19)are less noticeable than the respiratory symptoms,but they may be associated with disability and mortality in COVID-19.Even though Omicron caused less severe disease than Delta,the incidence of neurological manifestations is similar.More than 30%of patients experienced“brain fog”,delirium,stroke,and cognitive impairment,and over half of these patients presented abnormal neuroimaging outcomes.In this review,we summarize current advances in the clinical findings of neurological manifestations in COVID-19 patients and compare them with those in patients with influenza infection.We also illustrate the structure and cellular invasion mechanisms of SARS-CoV-2 and describe the pathway for central SARS-CoV-2 invasion.In addition,we discuss direct damage and other pathological conditions caused by SARS-CoV-2,such as an aberrant interferon response,cytokine storm,lymphopenia,and hypercoagulation,to provide treatment ideas.This review may offer new insights into preventing or treating brain damage in COVID-19.

NLRP3-mediated autophagy dysfunction links gut microbiota dysbiosis to tau pathology in chronic sleep deprivation

Emerging evidence indicates that sleep deprivation(SD)can lead to Alzheimer’s disease(AD)-related pathological changes and cognitive decline.However,the underlying mechanisms remain obscure.In the present study,we identified the existence of a microbiota-gut-brain axis in cognitive deficits resulting from chronic SD and revealed a potential pathway by which gut microbiota affects cognitive functioning in chronic SD.Our findings demonstrated that chronic SD in mice not only led to cognitive decline but also induced gut microbiota dysbiosis,elevated NLRP3 inflammasome expression,GSK-3βactivation,autophagy dysfunction,and tau hyperphosphorylation in the hippocampus.Colonization with the“SD microbiota”replicated the pathological and behavioral abnormalities observed in chronic sleep-deprived mice.Remarkably,both the deletion of NLRP3 in NLRP3-/-mice and specific knockdown of NLRP3 in the hippocampus restored autophagic flux,suppressed tau hyperphosphorylation,and ameliorated cognitive deficits induced by chronic SD,while GSK-3βactivity was not regulated by the NLRP3 inflammasome in chronic SD.Notably,deletion of NLRP3 reversed NLRP3 inflammasome activation,autophagy deficits,and tau hyperphosphorylation induced by GSK-3βactivation in primary hippocampal neurons,suggesting that GSK-3β,as a regulator of NLRP3-mediated autophagy dysfunction,plays a significant role in promoting tau hyperphosphorylation.Thus,gut microbiota dysbiosis was identified as a contributor to chronic SD-induced tau pathology via NLRP3-mediated autophagy dysfunction,ultimately leading to cognitive deficits.Overall,these findings highlight GSK-3βas a regulator of NLRP3-mediated autophagy dysfunction,playing a critical role in promoting tau hyperphosphorylation.

Clinical pathological characteristics of“crawling-type”gastric adenocarcinoma cancer:A case report

BACKGROUND Gastric cancer(GC)is a significant health problem worldwide,and early detection and accurate diagnosis are crucial for improving patient outcomes.Crawling-type gastric adenocarcinoma is a rare subtype of GC that has unique histopathological and clinical characteristics,and its diagnosis and management can be challenging.This pathological type of GC is also rare.CASE SUMMARY Here,we report the case of a patient who underwent ordinary endoscopy,na-rrow-band imaging,and endoscopic ultrasonography intending to determine the extent of tumor invasion and upper abdominal enhanced computed tomography and whether there was tumor metastasis.Then,endoscopic submucosal dissection was performed.After pathological and immunohistochemical examination,the pathological diagnosis was crawling-type gastric adenocarcinoma.This is a very rare and special pathological type of tumor.This case highlights the importance of using advanced endoscopic techniques and pathological examination in diagnosing and managing gastric crawling-type adenocarcinoma.Moreover,the findings underscore the need for continued research and clinical experience in this rare subtype of GC to improve patient outcomes.CONCLUSION The“crawling-type”GC is a rare and specific tumor pathology.It is difficult to identify and diagnose gliomas via endoscopy.The tumor is ill-defined,with a flat appearance and indistinct borders due to the lack of contrast against the background mucosa.Pathology revealed that the tumor cells were hand-like,so the patient has diagnosed with“crawling-type”gastric adenocarcinoma.

Digital pathology-based artificial intelligence models for differential diagnosis and prognosis of sporadic odontogenic keratocysts

Odontogenic keratocyst(OKC)is a common jaw cyst with a high recurrence rate.OKC combined with basal cell carcinoma as well as skeletal and other developmental abnormalities is thought to be associated with Gorlin syndrome.Moreover,OKC needs to be differentiated from orthokeratinized odontogenic cyst and other jaw cysts.Because of the different prognosis,differential diagnosis of several cysts can contribute to clinical management.We collected 519 cases,comprising a total of 2157 hematoxylin and eosinstained images,to develop digital pathology-based artificial intelligence(AI)models for the diagnosis and prognosis of OKC.The Inception_v3 neural network was utilized to train and test models developed from patch-level images.Finally,whole slide imagelevel AI models were developed by integrating deep learning-generated pathology features with several machine learning algorithms.The AI models showed great performance in the diagnosis(AUC=0.935,95%CI:0.898–0.973)and prognosis(AUC=0.840,95%CI:0.751–0.930)of OKC.The advantages of multiple slides model for integrating of histopathological information are demonstrated through a comparison with the single slide model.Furthermore,the study investigates the correlation between AI features generated by deep learning and pathological findings,highlighting the interpretative potential of AI models in the pathology.Here,we have developed the robust diagnostic and prognostic models for OKC.The AI model that is based on digital pathology shows promise potential for applications in odontogenic diseases of the jaw.

Going straight for the gut:gut-brain axis pathology and treatment of Parkinson's disease

This perspective focuses on the recent literature regarding the role of the gut-brain axis(GBA) in fecal microbiota transplantation(FMT) and stem cell therapy(SCT) in Parkinson's disease(PD).PD is the second most common neurodegenerative disease in the United States,yet therapies remain limited.Current research suggests that the GBA may play a role in the pathogenesis of PD.GBAbased FMT as well as SCT offer promising new avenues for PD treatment.Pro bing the interactions between FMT and SCT with the GBA may reveal novel therapeutics for PD.

Amyloid-beta pathology-induced nanoscale synaptic disruption:the case of the GABA_B-GIRK assembly

Alzheimer's disease (AD) is characterized by an imbalance between excitatory and inhibitory brain networks,leading to aberrant homeostatic synaptic plasticity.AD has progressively been recognized as syna ptopathy and syna ptic dysfunction has been identified as a key component of its pathogenesis (Schirinzi et al.,2020).Syna ptic dysfunction is believed to precede synapse loss,a primary biological correlate of cognitive decline in AD,inevita bly associated with neuronal death.

Clinical analysis of 12 cases of ovarian neuroendocrine carcinoma

BACKGROUND Neuroendocrine neoplasms of the female genital tract are rare.AIM To enhance our clinical understanding of neuroendocrine carcinoma(NEC)of the ovary.METHODS A retrospective review was conducted on 12 patients diagnosed with NEC of the ovary,analyzing clinicopathological characteristics,treatment modalities,and survival status.RESULTS The median age at diagnosis was 34.5 years(range:20 to 62 years).Among the 12 cases,9 were small cell carcinoma of the ovary and 3 were large cell NEC.Five cases were stage I tumors,one case was stage IV,and six cases were stage III.Eleven patients underwent surgery as part of their treatment.All patients received adjuvant chemotherapy.Among the 12 patients,one patient received radiotherapy,and one patient with a BRCA2 mutation was administered PARP inhibitor maintenance after chemotherapy.The median progression-free survival was 13 months,and the median overall survival was 19.5 months.Four cases remained disease-free,while eight cases experienced tumor recurrence,including three cases that resulted in death due to disease recurrence.CONCLUSION NEC of the ovary is a rare condition that is more common in women of childbearing age and is associated with aggressive behavior and poor clinical outcomes.Surgical resection remains the mainstay of treatment,with some patients benefiting from adjuvant chemoradiation therapy.

Hysterectomies for Gynaecological Pathology: 56 Cases at the Segou Regional Hospital in Mali

Introduction: Hysterectomy is a surgical procedure involving partial or total removal of the uterus. It is the most common gynaecological surgery in the world. Objective: To describe the epidemio-clinical and prognostic aspects of gynaecological hysterectomies. Patients and methods: This was an 18-month retrospective prospective descriptive study with a six-month follow-up period from 1 December 2020 to 31 May 2022 carried out in the gynaecology department of the Segou regional hospital. Results: Fifty-six (56) hysterectomies were performed out of 118 gynaecological surgical procedures (47.45%). The mean age was 47 ± 11.77 years. Large multiparous women were the most common (50%), with an average parity of 4.58. The main indications were uterine fibroids (30.4%), precancerous lesions of the cervix (17.85%) and uterine prolapse (17.85%). The abdominal route was the most commonly used surgical route (82.14%). Hysterectomy was total in 100% of cases and associated with bilateral adnexectomy in 48.2% of cases. The intra- and post-operative prognosis was satisfactory in 94.6% of cases. No deaths were recorded. The average length of stay was 3.28 days, irrespective of the surgical approach. Three cases of dyspareunia were noted among those who had resumed sexual activity.

Using MsfNet to Predict the ISUP Grade of Renal Clear Cell Carcinoma in Digital Pathology Images

Clear cell renal cell carcinoma(ccRCC)represents the most frequent form of renal cell carcinoma(RCC),and accurate International Society of Urological Pathology(ISUP)grading is crucial for prognosis and treatment selection.This study presents a new deep network called Multi-scale Fusion Network(MsfNet),which aims to enhance the automatic ISUP grade of ccRCC with digital histopathology pathology images.The MsfNet overcomes the limitations of traditional ResNet50 by multi-scale information fusion and dynamic allocation of channel quantity.The model was trained and tested using 90 Hematoxylin and Eosin(H&E)stained whole slide images(WSIs),which were all cropped into 320×320-pixel patches at 40×magnification.MsfNet achieved a micro-averaged area under the curve(AUC)of 0.9807,a macro-averaged AUC of 0.9778 on the test dataset.The Gradient-weighted Class Activation Mapping(Grad-CAM)visually demonstrated MsfNet’s ability to distinguish and highlight abnormal areas more effectively than ResNet50.The t-Distributed Stochastic Neighbor Embedding(t-SNE)plot indicates our model can efficiently extract critical features from images,reducing the impact of noise and redundant information.The results suggest that MsfNet offers an accurate ISUP grade of ccRCC in digital images,emphasizing the potential of AI-assisted histopathological systems in clinical practice.

A Multi-Task Deep Learning Framework for Simultaneous Detection of Thoracic Pathology through Image Classification

Thoracic diseases pose significant risks to an individual's chest health and are among the most perilous medical diseases. They can impact either one or both lungs, which leads to a severe impairment of a person’s ability to breathe normally. Some notable examples of such diseases encompass pneumonia, lung cancer, coronavirus disease 2019 (COVID-19), tuberculosis, and chronic obstructive pulmonary disease (COPD). Consequently, early and precise detection of these diseases is paramount during the diagnostic process. Traditionally, the primary methods employed for the detection involve the use of X-ray imaging or computed tomography (CT) scans. Nevertheless, due to the scarcity of proficient radiologists and the inherent similarities between these diseases, the accuracy of detection can be compromised, leading to imprecise or erroneous results. To address this challenge, scientists have turned to computer-based solutions, aiming for swift and accurate diagnoses. The primary objective of this study is to develop two machine learning models, utilizing single-task and multi-task learning frameworks, to enhance classification accuracy. Within the multi-task learning architecture, two principal approaches exist soft parameter sharing and hard parameter sharing. Consequently, this research adopts a multi-task deep learning approach that leverages CNNs to achieve improved classification performance for the specified tasks. These tasks, focusing on pneumonia and COVID-19, are processed and learned simultaneously within a multi-task model. To assess the effectiveness of the trained model, it is rigorously validated using three different real-world datasets for training and testing.