Multicenter integration analysis of TRP channels revealed potential mechanisms of immunosuppressive microenvironment activation and identified a machine learning-derived signature for improving outcomes in gliomas.

AIM: This study aimed to explore the mechanisms of transient receptor potential (TRP) channels on the immune microenvironment and develop a TRP-related signature for predicting prognosis, immunotherapy response, and drug sensitivity in gliomas. METHODS: Based on the unsupervised clustering algorithm, we identified novel TRP channel clusters and investigated their biological function, immune microenvironment, and genomic heterogeneity. In vitro and in vivo experiments revealed the association between TRPV2 and macrophages. Subsequently, based on 96 machine learning algorithms and six independent glioma cohorts, we constructed a machine learning-based TRP channel signature (MLTS). The performance of the MLTS in predicting prognosis, immunotherapy response, and drug sensitivity was evaluated. RESULTS: Patients with high expression levels of TRP channel genes had worse prognoses, higher tumor mutation burden, and more activated immunosuppressive microenvironment. Meanwhile, TRPV2 was identified as the most essential regulator in TRP channels. TRPV2 activation could promote macrophages migration toward malignant cells and alleviate glioma prognosis. Furthermore, MLTS could work independently of common clinical features and present stable and superior prediction performance. CONCLUSION: This study investigated the comprehensive effect of TRP channel genes in gliomas and provided a promising tool for designing effective, precise treatment strategies.

Construction of chitosan/alginate aerogels with three-dimensional hierarchical pore network structure via hydrogen bonding dissolution and covalent crosslinking synergistic strategy for thermal management systems.

To replace traditional petrochemical-based thermal insulation materials, in this work, the chitosan (CHI)/alginate (ALG) (CA) aerogels with three-dimensional hierarchical pore network structure were constructed by compositing CHI and ALG using a synergistic strategy of hydrogen bonding dissolution and covalent crosslinking. The structure and properties were further regulated by crosslinking the CA aerogels with epichlorohydrin (ECH). The CA aerogels exhibited various forms of covalent crosslinking, hydrogen bonding and electrostatic interactions, with hydrogen bonding content reaching 79.12 %. The CA aerogels showed an excellent three-dimensional hierarchical pore network structure, with an average pore size minimum of 15.92 nm. The structure regulation of CA aerogels obtained excellent compressive properties, with an increase of stress and strain by 137.61 % and 45.05 %, which can support a heavy object 5000 times its weight. Additionally, CA aerogels demonstrate excellent thermal insulation properties and low thermal conductivity, comparable to commercially available insulation materials. More importantly, CA aerogels have good cyclic insulation stability and thermal properties, and they have a flame retardancy rating of V-0, which shows the stability of insulation properties and excellent safety. CA aerogels provide new ideas for the development of biomass thermal insulation materials and are expected to be candidates for thermal management applications.

Exploring MAP2K3 as a prognostic biomarker and potential immunotherapy target in glioma treatment.

Glioma, the most prevalent primary brain tumor in adults, is characterized by significant invasiveness and resistance. Current glioma treatments include surgery, radiation, chemotherapy, and targeted therapy, but these methods often fail to eliminate the tumor completely, leading to recurrence and poor prognosis. Immune checkpoint inhibitors, a class of commonly used immunotherapeutic drugs, have demonstrated excellent efficacy in treating various solid malignancies. Recent research has indicated that unconventional levels of expression of the MAP2K3 gene closely correlates with glioma malignancy, hinting it could be a potential immunotherapy target. Our study unveiled substantial involvement of MAP2K3 in gliomas, indicating the potential of the enzyme to serve as a prognostic biomarker related to immunity. Through the regulation of the infiltration of immune cells, MAP2K3 can affect the prognosis of patients with glioma. These discoveries establish a theoretical foundation for exploring the biological mechanisms underlying MAP2K3 and its potential applications in glioma treatment.

Droplet drinking in constrictions.

Droplets generated through microfluidics serve as a common platform for assembling artificial cells, which are feasibly tailored using microfluidic methodology. The ability of natural cells to undergo shape changes, such as phagocytosis, is a typical characteristic that researchers aim to mimic in artificial cells. However, simulating the deformation behavior of natural cells within droplets is exceptionally challenging. Here, this study reports a pinocytosis-like phenomenon observed in droplets, termed "droplet drinking". When droplets traverse a capillary with constrictions, the shear force from the continuous-phase fluid induces relative motion within the droplets, creating concave regions at the rear. These regions facilitate engulfing of the continuous-phase fluid, resulting in the formation of multiple emulsions. This behavior is influenced by the capillary number, and the size of the ingested droplets is governed by the interfacial tension between the two phases. The production of multicore or multi-shell emulsions can be easily accomplished by making slight adjustments to the constrictions. Furthermore, this method demonstrates the integration of reactants into pre-existing droplets, facilitating biochemical reactions. This study presents a convenient approach for generating complex emulsions and an innovative strategy for studying deformation behavior in droplet-based artificial cells.

Mechanism of 7H-Dibenzo[c,g]carbazole metabolism in cytochrome P450 1A1: Insights from computational studies.

7H-Dibenzo[c,g]carbazole (DBC) is a prevalent environmental contaminant that induces tumorigenesis in several experimental animals. Recently, it has been utilized to develop high-performance solar cells and organic phosphorescent materials. It is imperative to strengthen investigations of DBC metabolism to understand its potential risks to human health. In this study, human CYP1A1 was employed as the metabolic enzyme to investigate the metabolic mechanism of DBC by molecular docking, molecular dynamics (MD) simulation, and quantum mechanical (QM) calculation. The results indicate that DBC binds to CYP1A1 in two modes (mode 1 and mode 2) mainly through nonpolar solvation energies (ΔGnonpolar). The formation of the two binding modes is attributed to the anchoring effect of the hydrogen bond formed by DBC with Asp320 (mode 1) or Ser116 (mode 2). Mode 1 is a "reactive" conformation, while mode 2 is not considered a "reactive" conformation. C5 is identified as the dominant site, and the pyrrole nitrogen cannot participate in the metabolism. DBC is metabolized mainly by a distinct electrophilic addition-rearrangement mechanism, with an energy barrier of 21.74 kcal/mol. The results provide meaningful insights into the biometabolic process of DBC and contribute to understanding its environmental effects and health risks.

Improving Buried Interface Contact by Bidentate Anchoring for Inverted Perovskite Solar Cells.

Nickel oxide (NiOx) is a promising hole transport layer (HTL) to fabricate efficient and large-scale inverted perovskite solar cells (PSCs) due to its low cost and superior chemical stability. However, inverted PSCs based on NiOx are still lagging behind that of other HTL because of the poor quality of buried interface contact. Herein, a bidentate ligand, 4,6-bis (diphenylphosphino) phenoxazine (2DPP), is used to regulate the NiOx surface and perovskite buried interface. The diphosphine Lewis base in the 2DPP molecule can coordinate both with NiOx and lead ions at NiOx/perovskite interface, leading to high-quality perovskite films with minimized defects. It is found that the inverted PSCs with 2DPP-modified buried interface exhibit double advantages of being both fast charge extraction and reduced nonradiative recombination, which is a combination of multiple factors including favorable energetic alignment, improved interface contact and strong binding between NiOx/2DPP and perovskite. The optimal PSC based on 2DPP modification yields a champion power conversion efficiency (PCE) of 21.9%. The unencapsulated PSC maintains above 75% of its initial PCE in the air with a relative humidity (RH) of 30-40% for 1000 h.

HepG2.2.15-derived exosomes facilitate the activation and fibrosis of hepatic stellate cells.

BACKGROUND: The role of exosomes derived from HepG2.2.15 cells, which express hepatitis B virus (HBV)-related proteins, in triggering the activation of LX2 liver stellate cells and promoting liver fibrosis and cell proliferation remains elusive. The focus was on comprehending the relationship and influence of differentially expressed microRNAs (DE-miRNAs) within these exosomes. AIM: To elucidate the effect of exosomes derived from HepG2.2.15 cells on the activation of hepatic stellate cell (HSC) LX2 and the progression of liver fibrosis. METHODS: Exosomes from HepG2.2.15 cells, which express HBV-related proteins, were isolated from parental HepG2 and WRL68 cells. Western blotting was used to confirm the presence of the exosomal marker protein CD9. The activation of HSCs was assessed using oil red staining, whereas DiI staining facilitated the observation of exosomal uptake by LX2 cells. Additionally, we evaluated LX2 cell proliferation and fibrosis marker expression using 5-ethynyl-2'-deoxyuracil staining and western blotting, respectively. DE-miRNAs were analyzed using DESeq2. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were used to annotate the target genes of DE-miRNAs. RESULTS: Exosomes from HepG2.2.15 cells were found to induced activation and enhanced proliferation and fibrosis in LX2 cells. A total of 27 miRNAs were differentially expressed in exosomes from HepG2.2.15 cells. GO analysis indicated that these DE-miRNA target genes were associated with cell differentiation, intracellular signal transduction, negative regulation of apoptosis, extracellular exosomes, and RNA binding. KEGG pathway analysis highlighted ubiquitin-mediated proteolysis, the MAPK signaling pathway, viral carcinogenesis, and the toll-like receptor signaling pathway, among others, as enriched in these targets. CONCLUSION: These findings suggest that exosomes from HepG2.2.15 cells play a substantial role in the activation, proliferation, and fibrosis of LX2 cells and that DE-miRNAs within these exosomes contribute to the underlying mechanisms.

Risk Factors of New Symptomatic Fractures After Vertebroplasty: A Retrospective Cohort Study of 268 Patients with Painful Osteoporotic Vertebral Compression Fracture.

OBJECTIVE: To evaluate the risk factors of new osteoporotic vertebral compression fractures (OVCFs) after percutaneous vertebroplasty (PVP). METHODS: From January 2016 to November 2019, patients suffering from OVCFs were retrospectively reviewed. The independent influence factors for new OVCFs after PVP were assessed, from following variables: age, sex, body mass index, bone mineral density (BMD), history of alcoholism, smoking, hypertension, diabetes, glucocorticoid use, and prior vertebral fractures, the number of initial fractures, mean cement volume, method of puncture, D-type of cement leakage, and regular antiosteoporosis treatment. RESULTS: A total of 268 patients with 347 levels met the inclusion criteria and were finally included in this study. Forty-nine levels of new OVCFs among 33 patients (12.31%) were observed during the follow-up period. It indicated that female (adjusted odds ratio [OR]: 6.812, 95% confidence interval {CI}: [1.096, 42.337], P = 0.040), lower BMD (adjusted OR: 0.477, 95% CI: [0.300, 0.759], P = 0.002), prior vertebral fractures (adjusted OR: 16.145, 95% CI: [5.319, 49.005], P = 0.000), and regular antiosteoporosis treatment (adjusted OR: 0.258, 95% CI: [0.086, 0.774], P = 0.016) were independent influence factors for new OVCF. The cut-off value of BMD to reach new OVCF was -3.350, with a sensitivity of 0.660 and a specificity of 0.848. CONCLUSION: Female, lower BMD (T-score of lumbar), prior vertebral fractures, and regular antiosteoporosis treatment were independent influencing factors. BMD (T-score of lumbar) lower than -3.350 would increase risk for new OVCF, and none osteoporotic treatment has detrimental effect on new onset fractures following PVP.

Antiviral and anti-inflammatory activities of chemical constituents from twigs of Mosla chinensis Maxim.

Seven undescribed compounds, including three flavones (1-3), one phenylpropanoid (19), three monoaromatic hydrocarbons (27-29), were isolated from the twigs of Mosla chinensis Maxim together with twenty-eight known compounds. The structures were characterized by HRESIMS, 1D and 2D NMR, and ECD spectroscopic techniques. Compound 20 displayed the most significant activity against A/WSN/33/2009 (H1N1) virus (IC50 = 20.47 µM) compared to the positive control oseltamivir (IC50 = 6.85 µM). Further research on the anti-influenza mechanism showed that compound 20 could bind to H1N1 virus surface antigen HA1 and inhibit the early attachment stage of the virus. Furthermore, compounds 9, 22, 23, and 25 displayed moderate inhibitory effects on the NO expression in LPS inducing Raw 264.7 cells with IC50 values of 22.78, 20.47, 27.66, and 30.14 µM, respectively.

A novel multidisciplinary machine learning approach based on clinical, imaging, colonoscopy, and pathology features for distinguishing intestinal tuberculosis from Crohn's disease.

OBJECTIVES: Differentiating intestinal tuberculosis (ITB) from Crohn's disease (CD) remains a diagnostic dilemma. Misdiagnosis carries potential grave implications. We aim to establish a multidisciplinary-based model using machine learning approach for distinguishing ITB from CD. METHODS: Eighty-two patients including 25 patients with ITB and 57 patients with CD were retrospectively recruited (54 in training cohort and 28 in testing cohort). The region of interest (ROI) for the lesion was delineated on magnetic resonance enterography (MRE) and colonoscopy images. Radiomic features were extracted by least absolute shrinkage and selection operator regression. Pathological feature was extracted automatically by deep-learning method. Clinical features were filtered by logistic regression analysis. Diagnostic performance was evaluated by receiver operating characteristic (ROC) curve and decision curve analysis (DCA). Delong's test was applied to compare the efficiency between the multidisciplinary-based model and the other four single-disciplinary-based models. RESULTS: The radiomics model based on MRE features yielded an AUC of 0.87 (95% confidence interval [CI] 0.68-0.96) on the test data set, which was similar to the clinical model (AUC, 0.90 [95% CI 0.71-0.98]) and higher than the colonoscopy radiomics model (AUC, 0.68 [95% CI 0.48-0.84]) and pathology deep-learning model (AUC, 0.70 [95% CI 0.49-0.85]). Multidisciplinary model, integrating 3 clinical, 21 MRE radiomic, 5 colonoscopy radiomic, and 4 pathology deep-learning features, could significantly improve the diagnostic performance (AUC of 0.94, 95% CI 0.78-1.00) on the bases of single-disciplinary-based models. DCA confirmed the clinical utility. CONCLUSIONS: Multidisciplinary-based model integrating clinical, MRE, colonoscopy, and pathology features was useful in distinguishing ITB from CD.

Mitochondria-associated endoplasmic reticulum membranes tethering protein VAPB-PTPIP51 protects against ischemic stroke through inhibiting the activation of autophagy.

AIMS: Mitochondria-associated endoplasmic reticulum membranes (MAMs) serve as a crucial bridge connecting the endoplasmic reticulum (ER) and mitochondria within cells. Vesicle-associated membrane protein-associated protein B (VAPB) and protein tyrosine phosphatase interacting protein 51 (PTPIP51) are responsible for the formation and stability of MAMs, which have been implicated in the pathogenesis of various diseases. However, the role of MAMs in ischemic stroke (IS) remains unclear. We aimed to investigate the role of MAMs tethering protein VAPB-PTPIP51 in experimental cerebral ischemia. METHODS: We simulated cerebral ischemia-reperfusion injury (CIRI) by using a mouse middle cerebral artery occlusion (MCAO) model. RESULTS: We observed a decrease in VAPB-PTPIP51 expression in the brain tissue. Our findings suggested compromised MAMs after MCAO, as a decreased mitochondria-ER contact (MERC) coverage and an increased distance were observed through the transmission electron microscope (TEM). Upon VAPB or PTPIP51 knockdown, the damage to MAMs was exacerbated, accompanied by excessive autophagy activation and increased reactive oxygen species (ROS) production, resulting in an enlarged infarct area and exacerbated neurological deficits. Notably, we observed that this damage was concomitant with the inhibition of the PI3K/AKT/mTOR pathway and was successfully mitigated by the treatment with the PI3K activator. CONCLUSIONS: Our findings suggest that the downregulation of VAPB-PTPIP51 expression after IS mediates structural damage to MAMs. This may exacerbate CIRI by inhibiting the PI3K pathway and activating autophagy, thus providing new therapeutic targets for IS.

Strategies for Treating Traumatic Neuromas with Tissue-Engineered Materials.

Neuroma, a pathological response to peripheral nerve injury, refers to the abnormal growth of nerve tissue characterized by disorganized axonal proliferation. Commonly occurring after nerve injuries, surgeries, or amputations, this condition leads to the formation of painful nodular structures. Traditional treatment options include surgical excision and pharmacological management, aiming to alleviate symptoms. However, these approaches often offer temporary relief without addressing the underlying regenerative challenges, necessitating the exploration of advanced strategies such as tissue-engineered materials for more comprehensive and effective solutions. In this study, we discussed the etiology, molecular mechanisms, and histological morphology of traumatic neuromas after peripheral nerve injury. Subsequently, we summarized and analyzed current nonsurgical and surgical treatment options, along with their advantages and disadvantages. Additionally, we emphasized recent advancements in treating traumatic neuromas with tissue-engineered material strategies. By integrating biomaterials, growth factors, cell-based approaches, and electrical stimulation, tissue engineering offers a comprehensive solution surpassing mere symptomatic relief, striving for the structural and functional restoration of damaged nerves. In conclusion, the utilization of tissue-engineered materials has the potential to significantly reduce the risk of neuroma recurrence after surgical treatment.

Integration analysis of cell division cycle-associated family genes revealed potential mechanisms of gliomagenesis and constructed an artificial intelligence-driven prognostic signature.

Cell division cycle-associated (CDCA) gene family members are essential cell proliferation regulators and play critical roles in various cancers. However, the function of the CDCA family genes in gliomas remains unclear. This study aims to elucidate the role of CDCA family members in gliomas using in vitro and in vivo experiments and bioinformatic analyses. We included eight glioma cohorts in this study. An unsupervised clustering algorithm was used to identify novel CDCA gene family clusters. Then, we utilized multi-omics data to elucidate the prognostic disparities, biological functionalities, genomic alterations, and immune microenvironment among glioma patients. Subsequently, the scRNA-seq analysis and spatial transcriptomic sequencing analysis were carried out to explore the expression distribution of CDCA2 in glioma samples. In vivo and in vitro experiments were used to investigate the effects of CDCA2 on the viability, migration, and invasion of glioma cells. Finally, based on ten machine-learning algorithms, we constructed an artificial intelligence-driven CDCA gene family signature called the machine learning-based CDCA gene family score (MLCS). Our results suggested that patients with the higher expression levels of CDCA family genes had a worse prognosis, more activated RAS signaling pathways, and more activated immunosuppressive microenvironments. CDCA2 knockdown inhibited the proliferation, migration, and invasion of glioma cells. In addition, the MLCS had robust and favorable prognostic predictive ability and could predict the response to immunotherapy and chemotherapy drug sensitivity.

Quercetin-induced pyroptosis in colon cancer through NEK7-mediated NLRP3 inflammasome-GSDMD signaling pathway activation.

Pyroptosis, a gasdermin-mediated lytic cell death, is a new hotspot topic in cancer research, and induction of tumor pyroptosis has emerged as a new target in cancer management. Quercetin (Que), a natural substance, demonstrates promising anticancer action. However, further information is required to fully comprehend the function and mechanism of Que in pyroptosis in colon cancer. This study revealed the underlying mechanism of Que-induced pyroptosis in colon cancer in vitro and in vivo. Que inhibited colon cancer cell growth through gasdermin D (GSDMD)-mediated pyroptosis. Depletion of GSDMD, rather than gasdermin E (GSDME), reversed the cytotoxic effects of Que on colon cancer cells. Que treatment upregulated NIMA-related kinase 7 (NEK7) protein expression, thus facilitating the assembly of the NLRP3 inflammasome and cleavage of GSDMD. NEK7 silencing resulted in colon cancer cell growth in vitro and in vivo. Mechanistically, NEK7 depression restrained the activation of the NLRP3 inflammasome-GSDMD pathway, thus attenuating pyroptosis triggered by Que in colon cancer cells. Furthermore, lower NEK7 and NLRP3 expression levels indicated colon cancer progression. Our results unveiled a novel pattern of anti-colon cancer activity of Que, and activation of NEK7-mediated pyroptosis is potentially a promising therapeutic target for colon cancer, which provides novel experimental proof for the clinical application of Que.

Hidden diversity and potential ecological function of phosphorus acquisition genes in widespread terrestrial bacteriophages.

Phosphorus (P) limitation of ecosystem processes is widespread in terrestrial habitats. While a few auxiliary metabolic genes (AMGs) in bacteriophages from aquatic habitats are reported to have the potential to enhance P-acquisition ability of their hosts, little is known about the diversity and potential ecological function of P-acquisition genes encoded by terrestrial bacteriophages. Here, we analyze 333 soil metagenomes from five terrestrial habitat types across China and identify 75 viral operational taxonomic units (vOTUs) that encode 105 P-acquisition AMGs. These AMGs span 17 distinct functional genes involved in four primary processes of microbial P-acquisition. Among them, over 60% (11/17) have not been reported previously. We experimentally verify in-vitro enzymatic activities of two pyrophosphatases and one alkaline phosphatase encoded by P-acquisition vOTUs. Thirty-six percent of the 75 P-acquisition vOTUs are detectable in a published global topsoil metagenome dataset. Further analyses reveal that, under certain circumstances, the identified P-acquisition AMGs have a greater influence on soil P availability and are more dominant in soil metatranscriptomes than their corresponding bacterial genes. Overall, our results reinforce the necessity of incorporating viral contributions into biogeochemical P cycling.

Predicting One-year Recurrence of HCC based on Gadoxetic Acid-enhanced MRI by Machine Learning Models.

OBJECTIVE: Accurate prediction of recurrence risk after resction in patients with Hepatocellular Carcinoma (HCC) may help to individualize therapy strategies. This study aimed to develop machine learning models based on preoperative clinical factors and multiparameter Magnetic Resonance Imaging (MRI) characteristics to predict the 1-year recurrence after HCC resection. METHODS: Eighty-two patients with single HCC who underwent surgery were retrospectively analyzed. All patients underwent preoperative gadoxetic acidenhanced MRI examination. Preoperative clinical factors and MRI characteristics were collected for feature selection. Least Absolute Shrinkage and Selection Operator (LASSO) was applied to select the optimal features for predicting postoperative 1-year recurrence of HCC. Four machine learning algorithms, Multilayer Perception (MLP), random forest, support vector machine, and k-nearest neighbor, were used to construct the predictive models based on the selected features. A Receiver Operating Characteristic (ROC) curve was used to assess the performance of each model. RESULTS: Among the enrolled patients, 32 patients experienced recurrences within one year, while 50 did not. Tumor size, peritumoral hypointensity, decreasing ratio of liver parenchyma T1 value (ΔT1), and α-fetoprotein (AFP) levels were selected by using LASSO to develop the machine learning models. The area under the curve (AUC) of each model exceeded 0.72. Among the models, the MLP model showed the best performance with an AUC, accuracy, sensitivity, and specificity of 0.813, 0.742, 0.570, and 0.853, respectively. CONCLUSION: Machine learning models can accurately predict postoperative 1-year recurrence in patients with HCC, which may help to provide individualized treatment.

Filament formation drives catalysis by glutaminase enzymes important in cancer progression.

The glutaminase enzymes GAC and GLS2 catalyze the hydrolysis of glutamine to glutamate, satisfying the 'glutamine addiction' of cancer cells. They are the targets of anti-cancer drugs; however, their mechanisms of activation and catalytic activity have been unclear. Here we demonstrate that the ability of GAC and GLS2 to form filaments is directly coupled to their catalytic activity and present their cryo-EM structures which provide a view of the conformational states essential for catalysis. Filament formation guides an 'activation loop' to assume a specific conformation that works together with a 'lid' to close over the active site and position glutamine for nucleophilic attack by an essential serine. Our findings highlight how ankyrin repeats on GLS2 regulate enzymatic activity, while allosteric activators stabilize, and clinically relevant inhibitors block, filament formation that enables glutaminases to catalyze glutaminolysis and support cancer progression.

Deep learning-based accurate diagnosis and quantitative evaluation of microvascular invasion in hepatocellular carcinoma on whole-slide histopathology images.

BACKGROUND: Microvascular invasion (MVI) is an independent prognostic factor that is associated with early recurrence and poor survival after resection of hepatocellular carcinoma (HCC). However, the traditional pathology approach is relatively subjective, time-consuming, and heterogeneous in the diagnosis of MVI. The aim of this study was to develop a deep-learning model that could significantly improve the efficiency and accuracy of MVI diagnosis. MATERIALS AND METHODS: We collected H&E-stained slides from 753 patients with HCC at the First Affiliated Hospital of Zhejiang University. An external validation set with 358 patients was selected from The Cancer Genome Atlas database. The deep-learning model was trained by simulating the method used by pathologists to diagnose MVI. Model performance was evaluated with accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve. RESULTS: We successfully developed a MVI artificial intelligence diagnostic model (MVI-AIDM) which achieved an accuracy of 94.25% in the independent external validation set. The MVI positive detection rate of MVI-AIDM was significantly higher than the results of pathologists. Visualization results demonstrated the recognition of micro MVIs that were difficult to differentiate by the traditional pathology. Additionally, the model provided automatic quantification of the number of cancer cells and spatial information regarding MVI. CONCLUSIONS: We developed a deep learning diagnostic model, which performed well and improved the efficiency and accuracy of MVI diagnosis. The model provided spatial information of MVI that was essential to accurately predict HCC recurrence after surgery.

Development and validation of a deep learning radiomics model with clinical-radiological characteristics for the identification of occult peritoneal metastases in patients with pancreatic ductal adenocarcinoma.

BACKGROUND: Occult peritoneal metastases (OPM) in patients with pancreatic ductal adenocarcinoma (PDAC) are frequently overlooked during imaging. The authors aimed to develop and validate a computed tomography (CT)-based deep learning-based radiomics (DLR) model to identify OPM in PDAC before treatment. METHODS: This retrospective, bicentric study included 302 patients with PDAC (training: n =167, OPM-positive, n =22; internal test: n =72, OPM-positive, n =9: external test, n =63, OPM-positive, n =9) who had undergone baseline CT examinations between January 2012 and October 2022. Handcrafted radiomics (HCR) and DLR features of the tumor and HCR features of peritoneum were extracted from CT images. Mutual information and least absolute shrinkage and selection operator algorithms were used for feature selection. A combined model, which incorporated the selected clinical-radiological, HCR, and DLR features, was developed using a logistic regression classifier using data from the training cohort and validated in the test cohorts. RESULTS: Three clinical-radiological characteristics (carcinoembryonic antigen 19-9 and CT-based T and N stages), nine HCR features of the tumor, 14 DLR features of the tumor, and three HCR features of the peritoneum were retained after feature selection. The combined model yielded satisfactory predictive performance, with an area under the curve (AUC) of 0.853 (95% CI: 0.790-0.903), 0.845 (95% CI: 0.740-0.919), and 0.852 (95% CI: 0.740-0.929) in the training, internal test, and external test cohorts, respectively (all P <0.05). The combined model showed better discrimination than the clinical-radiological model in the training (AUC=0.853 vs. 0.612, P <0.001) and the total test (AUC=0.842 vs. 0.638, P <0.05) cohorts. The decision curves revealed that the combined model had greater clinical applicability than the clinical-radiological model. CONCLUSIONS: The model combining CT-based DLR and clinical-radiological features showed satisfactory performance for predicting OPM in patients with PDAC.

An integrated proteomics and metabolomics approach to assess graft quality and predict early allograft dysfunction after liver transplantation: a retrospective cohort study.

BACKGROUND: Early allograft dysfunction (EAD) is a common complication after liver transplantation (LT) and is associated with poor prognosis. Graft itself plays a major role in the development of EAD. We aimed to reveal the EAD-specific molecular profiles to assess graft quality and establish EAD predictive models. METHODS: A total of 223 patients who underwent LT were enrolled and divided into training ( n =73) and validation ( n =150) sets. In the training set, proteomics was performed on graft biopsies, together with metabolomics on paired perfusates. Differential expression, enrichment analysis, and protein-protein interaction network were used to identify the key molecules and pathways involved. EAD predictive models were constructed using machine learning and verified in the validation set. RESULTS: A total of 335 proteins were differentially expressed between the EAD and non-EAD groups. These proteins were significantly enriched in triglyceride and glycerophospholipid metabolism, neutrophil degranulation, and the MET-related signaling pathway. The top 12 graft proteins involved in the aforementioned processes were identified, including GPAT1, LPIN3, TGFB1, CD59, and SOS1. Moreover, downstream metabolic products, such as lactate dehydrogenase, interleukin-8, triglycerides, and the phosphatidylcholine/phosphorylethanolamine ratio in the paired perfusate displayed a close relationship with the graft proteins. To predict the occurrence of EAD, an integrated model using perfusate metabolic products and clinical parameters showed areas under the curve of 0.915 and 0.833 for the training and validation sets, respectively. It displayed superior predictive efficacy than that of currently existing models, including donor risk index and D-MELD scores. CONCLUSIONS: We identified novel biomarkers in both grafts and perfusates that could be used to assess graft quality and provide new insights into the etiology of EAD. Herein, we also offer a valid tool for the early prediction of EAD.