Super-resolution diffractive neural network for all-optical direction of arrival estimation beyond diffraction limits.

Wireless sensing of the wave propagation direction from radio sources lays the foundation for communication, radar, navigation, etc. However, the existing signal processing paradigm for the direction of arrival estimation requires the radio frequency electronic circuit to demodulate and sample the multichannel baseband signals followed by a complicated computing process, which places the fundamental limit on its sensing speed and energy efficiency. Here, we propose the super-resolution diffractive neural networks (S-DNN) to process electromagnetic (EM) waves directly for the DOA estimation at the speed of light. The multilayer meta-structures of S-DNN generate super-oscillatory angular responses in local angular regions that can perform the all-optical DOA estimation with angular resolutions beyond the diffraction limit. The spatial-temporal multiplexing of passive and reconfigurable S-DNNs is utilized to achieve high-resolution DOA estimation over a wide field of view. The S-DNN is validated for the DOA estimation of multiple radio sources over 5 GHz frequency bandwidth with estimation latency over two to four orders of magnitude lower than the state-of-the-art commercial devices in principle. The results achieve the angular resolution over an order of magnitude, experimentally demonstrated with four times, higher than diffraction-limited resolution. We also apply S-DNN's edge computing capability, assisted by reconfigurable intelligent surfaces, for extremely low-latency integrated sensing and communication with low power consumption. Our work is a significant step towards utilizing photonic computing processors to facilitate various wireless sensing and communication tasks with advantages in both computing paradigms and performance over electronic computing.

Prospective Comparison of 2D and 3D T2-Weighted Imaging in Multiparametric MRI for Assessing Muscle Invasion Accuracy Using VI-RADS in Bladder Cancer.

RATIONALE AND OBJECTIVES: T2-weighted imaging (T2WI) is an essential sequence for assessing the staging of bladder cancer. This study aimed to compare the image quality and diagnostic performance of three-dimensional (3D) and two-dimensional (2D) T2WI in diagnosing muscle invasion of bladder cancer using Vesical Imaging Reporting and Data System (VI-RADS). MATERIALS AND METHODS: Between August 2022 and May 2023, 101 participants with bladder cancer underwent multiparametric MRI including 3D and 2D T2WI. Two radiologists independently reviewed 2D and 3D T2WI, evaluating image quality and muscle invasion based on VI-RADS scoring. The paired Wilcoxon signed-rank test assessed the differences between 2D and 3D T2WI. The areas under the receiver operating characteristic curve (AUCs) were utilized to compare the diagnostic performance. RESULTS: 3D T2WI demonstrated significantly superior overall image quality scores with less artifacts than 2D T2WI. Compared to 2D T2WI, 3D T2WI categories had significantly higher AUC for both readers (reader 1: 0.937 vs. 0.909, p = .02; reader 2: 0.923 vs.0.884, p = .04). The VI-RADS score of 3D MR protocol had higher accuracy than 2D MR protocol (reader 1: 0.931 vs. 0.921, p = .02; reader 2: 0.931 vs. 0.911, p = .02). However, there were no significant differences in AUC values of VI-RADS categories between 2D and 3D MR protocol (all p > 0.05). CONCLUSION: In assessing muscle invasion of bladder cancer, 3D T2WI exhibited superior overall image quality and diagnostic performance than 2D T2WI. However, 3D T2WI did not significantly improve the diagnostic performance of VI-RADS.

Pancreatic stellate cells and the interleukin family: Linking fibrosis and immunity to pancreatic ductal adenocarcinoma (Review).

Pancreatic ductal adenocarcinoma (PDAC) is an extremely aggressive form of cancer with a low survival rate. A successful treatment strategy should not be limited to targeting cancer cells alone, but should adopt a more comprehensive approach, taking into account other influential factors. These include the extracellular matrix (ECM) and immune microenvironment, both of which are integral components of the tumor microenvironment. The present review describes the roles of pancreatic stellate cells, differentiated cancer­associated fibroblasts and the interleukin family, either independently or in combination, in the progression of precursor lesions in pancreatic intraepithelial neoplasia and PDAC. These elements contribute to ECM deposition and immunosuppression in PDAC. Therapeutic strategies that integrate interleukin and/or stromal blockade for PDAC immunomodulation and fibrogenesis have yielded inconsistent results. A deeper comprehension of the intricate interplay between fibrosis, and immune responses could pave the way for more effective treatment targets, by elucidating the mechanisms and causes of ECM fibrosis during PDAC progression.

Proprotein Convertase Subtilisin/Kexin Type 9 Inhibitor Improves the Vascular Function of Arteriovenous Fistula in Rats with Hyperglycemia.

Objectives: Few evidence-based medications to improve the primary patency of arteriovenous fistulas in patients with diabetes who require hemodialysis are available. We investigated whether proprotein convertase subtilisin/kexin type 9 inhibitors (PCSK9i) could improve arteriovenous fistula function through pleiotropic effects in a rat model of hyperglycemia. Methods: Ex vivo effects of PCSK9i on the aorta of Sprague-Dawley (SD) rats were investigated using an organ bath system. For in vivo experiments, an abdominal aortocaval (AC) fistula was generated in SD rats (200-250 g) after inducing hyperglycemia through streptozotocin administration (80 mg/kg, intraperitoneal). Alirocumab (50 mg/kg/week, subcutaneous) was administered on the day of fistula surgery and day 7. Echocardiography, blood flow through the aorta-limb, vasomotor reactivity, and serum biochemistry were examined on D14. Furthermore, enzyme-linked immunosorbent assay and immunoblotting were performed. Results: PCSK9i induced aorta relaxation ex vivo through a potassium channel-associated mechanism. PCSK9i significantly improved blood flow and preserved endothelial function without changes in cardiac function and serum lipid levels in rats with hyperglycemia. The levels of lectin-like oxidized low-density lipoprotein receptor-1, superoxide dismutase, cyclooxygenase-2, caspase-1, and interleukin-1ß were significantly reduced in the treatment group. PCSK9i decreased the ratio of phosphorylated to total p38 mitogen-activated protein kinase and extracellular signal-regulated kinase in the aorta of rats with hyperglycemia. Conclusions: Short-term treatment with PCSK9i preserved endothelial function, induced vascular dilatation, and increased blood flow in the AC fistula of rats with hyperglycemia. The pleiotropic mechanisms were associated with the suppression of oxidative stress and tissue inflammation during hyperglycemia.

Self-Explainable Graph Neural Network for Alzheimer Disease and Related Dementias Risk Prediction: Algorithm Development and Validation Study.

BACKGROUND: Alzheimer disease and related dementias (ADRD) rank as the sixth leading cause of death in the United States, underlining the importance of accurate ADRD risk prediction. While recent advancements in ADRD risk prediction have primarily relied on imaging analysis, not all patients undergo medical imaging before an ADRD diagnosis. Merging machine learning with claims data can reveal additional risk factors and uncover interconnections among diverse medical codes. OBJECTIVE: The study aims to use graph neural networks (GNNs) with claim data for ADRD risk prediction. Addressing the lack of human-interpretable reasons behind these predictions, we introduce an innovative, self-explainable method to evaluate relationship importance and its influence on ADRD risk prediction. METHODS: We used a variationally regularized encoder-decoder GNN (variational GNN [VGNN]) integrated with our proposed relation importance method for estimating ADRD likelihood. This self-explainable method can provide a feature-important explanation in the context of ADRD risk prediction, leveraging relational information within a graph. Three scenarios with 1-year, 2-year, and 3-year prediction windows were created to assess the model's efficiency, respectively. Random forest (RF) and light gradient boost machine (LGBM) were used as baselines. By using this method, we further clarify the key relationships for ADRD risk prediction. RESULTS: In scenario 1, the VGNN model showed area under the receiver operating characteristic (AUROC) scores of 0.7272 and 0.7480 for the small subset and the matched cohort data set. It outperforms RF and LGBM by 10.6% and 9.1%, respectively, on average. In scenario 2, it achieved AUROC scores of 0.7125 and 0.7281, surpassing the other models by 10.5% and 8.9%, respectively. Similarly, in scenario 3, AUROC scores of 0.7001 and 0.7187 were obtained, exceeding 10.1% and 8.5% than the baseline models, respectively. These results clearly demonstrate the significant superiority of the graph-based approach over the tree-based models (RF and LGBM) in predicting ADRD. Furthermore, the integration of the VGNN model and our relation importance interpretation could provide valuable insight into paired factors that may contribute to or delay ADRD progression. CONCLUSIONS: Using our innovative self-explainable method with claims data enhances ADRD risk prediction and provides insights into the impact of interconnected medical code relationships. This methodology not only enables ADRD risk modeling but also shows potential for other image analysis predictions using claims data.

Vaginal microbial profile of cervical cancer patients receiving chemoradiotherapy: the potential involvement of Lactobacillus iners in recurrence.

The vaginal microbiome is an immune defense against reproductive diseases and can serve as an important biomarker for cervical cancer. However, the intrinsic relationship between the recurrence and the vaginal microbiome in patients with cervical cancer before and after concurrent chemoradiotherapy is poorly understood. Here, we analyzed 125 vaginal microbial profiles from a patient cohort of stage IB-IVB cervical cancer using 16S metagenomic sequencing and deciphered the microbial composition and functional characteristics of the recurrent and non-recurrent both before and after chemoradiotherapy. We demonstrated that the abundance of beneficial bacteria and stability of the microbial community in the vagina decreased in the recurrence group, implying the unique characteristics of the vaginal microbiome for recurrent cervical cancer. Moreover, using machine learning, we identified Lactobacillus iners as the most important biomarker, combined with age and other biomarkers (such as Ndongobacter massiliensis, Corynebacterium pyruviciproducens ATCC BAA-1742, and Prevotella buccalis), and could predict cancer recurrence phenotype before chemoradiotherapy. This study prospectively employed rigorous bioinformatics analysis and highlights the critical role of vaginal microbiota in post-treatment cervical cancer recurrence, identifying promising biomarkers with prognostic significance in the context of concurrent chemoradiotherapy for cervical cancer. The role of L. iners in determining chemoradiation resistance in cervical cancer warrants further detailed investigation. Our results expand our understanding of cervical cancer recurrence and help develop better strategies for prognosis prediction and personalized therapy.

Host-defence caerin 1.1 and 1.9 peptides suppress glioblastoma U87 and U118 cell proliferation through the modulation of mitochondrial respiration and induce the downregulation of CHI3L1.

Glioblastoma, the most aggressive form of brain cancer, poses a significant global health challenge with a considerable mortality rate. With the predicted increase in glioblastoma incidence, there is an urgent need for more effective treatment strategies. In this study, we explore the potential of caerin 1.1 and 1.9, host defence peptides derived from an Australian tree frog, in inhibiting glioblastoma U87 and U118 cell growth. Our findings demonstrate the inhibitory impact of caerin 1.1 and 1.9 on cell growth through CCK8 assays. Additionally, these peptides effectively curtail the migration of glioblastoma cells in a cell scratch assay, exhibiting varying inhibitory effects among different cell lines. Notably, the peptides hinder the G0/S phase replication in both U87 and U118 cells, pointing to their impact on the cell cycle. Furthermore, caerin 1.1 and 1.9 show the ability to enter the cytoplasm of glioblastoma cells, influencing the morphology of mitochondria. Proteomics experiments reveal intriguing insights, with a decrease in CHI3L1 expression and an increase in PZP and JUNB expression after peptide treatment. These proteins play roles in cell energy metabolism and inflammatory response, suggesting a multifaceted impact on glioblastoma cells. In conclusion, our study underscores the substantial anticancer potential of caerin 1.1 and 1.9 against glioblastoma cells. These findings propose the peptides as promising candidates for further exploration in the realm of glioblastoma management, offering new avenues for developing effective treatment strategies.

Click Chemistry-Mediated Polymannose Surface-Engineering of Natural Killer Cells for Immunotherapy of Triple-Negative Breast Cancer.

Natural killer (NK) cells, serve as the frontline defense of the immune system, and are capable of surveilling and eliminating tumor cells. Their significance in tumor immunotherapy has garnered considerable attention in recent years. However, the absence of specific receptor-ligand interactions between NK cells and tumor cells hampers their selectivity, thereby limiting the therapeutic effectiveness of NK cell-based tumor immunotherapy. Herein, this work constructs polymannose-engineered NK (pM-NK) cells via metabolic glycoengineering and copper-free click chemistry. Polymannose containing dibenzocyclooctyne terminal groups (pM-DBCO) is synthesized and covalently modified on the surface of azido-labeled NK cells. Compared to the untreated NK cells, the interactions between pM-NK cells and MDA-MB-231 cells, a breast tumor cell line with overexpression of mannose receptors (MRs), are significantly increased, and lead to significantly enhanced killing efficacy. Consequently, intravenous administration of pM-NK cells will effectively inhibit the tumor growth and will prolong the survival of mice bearing MDA-MB-231 tumors. Thus, this work presents a novel strategy for tumor-targeting NK cell-based tumor immunotherapy.

Reduced field-of-view DWI based on deep learning reconstruction improving diagnostic accuracy of VI-RADS for evaluating muscle invasion.

OBJECTIVES: To investigate whether reduced field-of-view (rFOV) diffusion-weighted imaging (DWI) with deep learning reconstruction (DLR) can improve the accuracy of evaluating muscle invasion using VI-RADS. METHODS: Eighty-six bladder cancer participants who were evaluated by conventional full field-of-view (fFOV) DWI, standard rFOV (rFOVSTA) DWI, and fast rFOV with DLR (rFOVDLR) DWI were included in this prospective study. Tumors were categorized according to the vesical imaging reporting and data system (VI-RADS). Qualitative image quality scoring, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and ADC value were evaluated. Friedman test with post hoc test revealed the difference across the three DWIs. Receiver operating characteristic analysis was performed to calculate the areas under the curve (AUCs). RESULTS: The AUC of the rFOVSTA DWI and rFOVDLR DWI were higher than that of fFOV DWI. rFOVDLR DWI reduced the acquisition time from 5:02 min to 3:25 min, and showed higher scores in overall image quality with higher CNR and SNR, compared to rFOVSTA DWI (p < 0.05). The mean ADC of all cases of rFOVSTA DWI and rFOVDLR DWI was significantly lower than that of fFOV DWI (all p < 0.05). There was no difference in mean ADC value and the AUC for evaluating muscle invasion between rFOVSTA DWI and rFOVDLR DWI (p > 0.05). CONCLUSIONS: rFOV DWI with DLR can improve the diagnostic accuracy of fFOV DWI for evaluating muscle invasion. Applying DLR to rFOV DWI reduced the acquisition time and improved overall image quality while maintaining ADC value and diagnostic accuracy. CRITICAL RELEVANCE STATEMENT: The diagnostic performance and image quality of full field-of-view DWI, reduced field-of-view (rFOV) DWI with and without DLR were compared. DLR would benefit the wide clinical application of rFOV DWI by reducing the acquisition time and improving the image quality. KEY POINTS: Deep learning reconstruction (DLR) can reduce scan time and improve image quality. Reduced field-of-view (rFOV) diffusion-weighted imaging (DWI) with DLR showed better diagnostic performances than full field-of-view DWI. There was no difference of diagnostic accuracy between rFOV DWI with DLR and standard rFOV DWI.

Mechanistic investigation of wogonin in delaying the progression of endothelial mesenchymal transition by targeting the TGF-ß1 pathway in pulmonary hypertension.

Pulmonary hypertension (PH) is characterized by pulmonary vascular remodeling, which endothelial-to-mesenchymal transition (EndMT) being its main progressive phase. Wogonin, a flavonoid extracted from the root of Scutellaria baicalensis Georgi, hinders the abnormal proliferation of cells and has been employed in the treatment of several cardiopulmonary diseases. This study was designed to investigate how wogonin affected EndMT during PH. Monocrotaline (MCT) was used to induce PH in rats. Binding capacity of TGF-ß1 receptor to wogonin detected by molecular docking and molecular dynamics. EndMT model was established in pulmonary microvascular endothelial cells (PMVECs) by transforming growth factor beta-1 (TGF-ß1). The result demonstrated that wogonin (20 mg/kg/day) attenuated right ventricular systolic pressure (RVSP), right ventricular hypertrophy and pulmonary vascular thickness in PH rats. EndMT in the pulmonary vascular was inhibited after wogonin treatment as evidenced by the restored expression of CD31 and decreased expression of α-SMA. Wogonin has strong affinity for both TGFBRI and TGFBRII, and has a better binding stability for TGFBRI. In TGF-ß1-treated PMVECs, wogonin (0.3, 1, and 3 µM) exhibited significant inhibitory effects on this transformation process via down-regulating the expression of p-Smad2 and Snail, while up-regulating the expression of p-Smad1/5. Additionally, results of Western blot and fluorescence shown that the expression of α-SMA were decrease with increasing level of CD31 in PMVECs. In conclusion, our research showed that wogonin suppressed EndMT via the TGF-ß1/Smad pathway which may lead to its alleviated effect on PH. Wogonin may be a promising drug against PH.

Cytotoxicity effect and transcriptome analysis of PCV3-infected cells revealed potential viral pathogenic mechanisms.

Porcine circovirus type 3 (PCV3) has become an important pathogen in the global swine industry and poses a threat to pig health, but its pathogenic mechanism remains unknown. In this study, we constructed an innovative, linear infectious clone of PCV3 for rescuing the virus, and explored the transcriptome of infected cells to gain insights into its pathogenic mechanisms. Subsequently, an in vivo experiment was conducted to evaluate the pathogenicity of the rescued virus in pig. PCV3 nucleic acid was distributed across various organs, indicating systemic circulation via the bloodstream and viremia. Immunohistochemical staining also revealed a significant presence of PCV3 antigens in the spleen, lungs, and lymph nodes, indicating that PCV3 had tropism for these organs. Transcriptome analysis of infected ST cells revealed differential expression of genes associated with apoptosis, immune responses, and cellular metabolism. Notably, upregulation of genes related to the hypoxia-inducible factor-1 pathway, glycolysis, and the AGE/RAGE pathway suggests activation of inflammatory responses, ultimately leading to onset of disease. These findings have expanded our understanding of PCV3 pathogenesis, and the interplay between PCV3 and host factors.

A Prospective Randomized Controlled Trial Assessing the Impact of Preoperative Combined with Postoperative Progressive Resistance Training on Muscle Strength, Gait, Balance and Function in Patients Undergoing Total Hip Arthroplasty.

Purpose: The aim of this study is to investigate the effects of a preoperative combined with postoperative moderate-intensity progressive resistance training (PRT) of the operative side in patients with hip osteoarthritis (HOA) who are undergoing total hip arthroplasty (THA). The study seeks to evaluate the impact of this combined intervention on muscle strength, gait, balance, and hip joint function in a controlled, measurable, and objective manner. Additionally, the study aims to compare the outcomes of this combined intervention with those of preoperative or postoperative muscle strength training conducted in isolation. Methods: A total of 90 patients with HOA scheduled for unilateral primary THA were randomly assigned to three groups: Pre group (preoperative PRT), Post group (postoperative PRT), and Pre& Post group (preoperative combined with postoperative PRT) focusing on hip flexion, extension, adduction, and abduction of operated side. Muscle strength, gait parameters, balance, and hip function were assessed at specific time points during a 12-month follow-up period. Results: All three groups showed significant improvements in muscle strength, with the Pre& Post group demonstrating the most pronounced and sustained gains. Gait velocity and cadence were significantly improved in the Pre& Post group at 1-month and 3-month postoperative follow-ups compared to the other groups. Similarly, the Pre& Post group exhibited superior balance performance at 3-month and 12-month postoperative follow-ups. The Harris Hip Score also showed better outcomes in the Pre& Post group at all follow-up intervals. Conclusion: Preoperative combined with postoperative moderate-intensity PRT in HOA patients undergoing THA led to superior improvements in muscle strength, gait, balance, and hip joint function compared to preoperative or postoperative PRT alone. This intervention shows significant promise in optimizing postoperative rehabilitation and enhancing patient outcomes following THA.

Deficiency of P2RY11 causes narcolepsy and attenuates the recruitment of neutrophils and macrophages in the inflammatory response in zebrafish.

Purinergic receptor P2Y11, a G protein-coupled receptor that is stimulated by extracellular ATP, has been demonstrated to be related to the chemotaxis of granulocytes, apoptosis of neutrophils, and secretion of cytokines in vitro. P2Y11 mutations were associated with narcolepsy. However, little is known about the roles of P2RY11 in the occurrence of narcolepsy and inflammatory response in vivo. In this study, we generated a zebrafish P2Y11 mutant using CRISPR/Cas9 genome editing and demonstrated that the P2Y11 mutant replicated the narcolepsy-like features including reduced HCRT expression and excessive daytime sleepiness, suggesting that P2Y11 is essential for HCRT expression. Furthermore, we accessed the cytokine expression in the mutant and revealed that the P2RY11 mutation disrupted the systemic inflammatory balance by reducing il4, il10 and tgfb, and increasing il6, tnfa, and il1b. In addition, the P2RY11-deficient larvae with caudal fin injuries exhibited significantly slower migration and less recruitment of neutrophils and macrophages at damaged site, and lower expression of anti-inflammatory cytokines during tissue damage. All these findings highlight the vital roles of P2RY11 in maintaining HCRT production and secreting anti-inflammatory cytokines in the native environment, and suggested that P2RY11-deficient zebrafish can serve as a reliable and unique model to further explore narcolepsy and inflammatory-related diseases with impaired neutrophil and macrophage responses.

Impact of PEDV infection on the biological characteristics of porcine intestinal exosomes.

Porcine epidemic diarrhea (PED) is a highly contagious intestinal infection primarily affecting pigs. It is caused by the porcine epidemic diarrhea virus (PEDV). PEDV targets the villus tissue cells in the small intestine and mesenteric lymph nodes, resulting in shortened intestinal villi and, in extreme cases, causing necrosis of the intestinal lining. Moreover, PEDV infection can disrupt the balance of the intestinal microflora, leading to an overgrowth of harmful bacteria like Escherichia coli. Exosomes, tiny membrane vesicles ranging from 30 to 150 nm in size, contain a complex mixture of RNA and proteins. MicroRNA (miRNA) regulates various cell signaling, development, and disease progression processes. This study extracted exosomes from both groups and performed high-throughput miRNA sequencing and bioinformatics techniques to investigate differences in miRNA expression within exosomes isolated from PEDV-infected porcine small intestine tissue compared to healthy controls. Notably, two miRNA types displayed upregulation in infected exosomes, while 12 exhibited downregulation. These findings unveil abnormal miRNA regulation patterns in PEDV-infected intestinal exosomes, shedding light on the intricate interplay between PEDV and its host. This will enable further exploration of the relationship between these miRNA changes and signaling pathways, enlightening PEDV pathogenesis and potential therapeutic targets.

Recurrent RhoGAP gene fusion CLDN18-ARHGAP26 promotes RHOA activation and focal adhesion kinase and YAP-TEAD signalling in diffuse gastric cancer.

OBJECTIVE: Genomic studies of gastric cancer have identified highly recurrent genomic alterations impacting RHO signalling, especially in the diffuse gastric cancer (DGC) histological subtype. Among these alterations are interchromosomal translations leading to the fusion of the adhesion protein CLDN18 and RHO regulator ARHGAP26. It remains unclear how these fusion constructs impact the activity of the RHO pathway and what is their broader impact on gastric cancer development. Herein, we developed a model to allow us to study the function of this fusion protein in the pathogenesis of DGC and to identify potential therapeutic targets for DGC tumours with these alterations. DESIGN: We built a transgenic mouse model with LSL-CLDN18-ARHGAP26 fusion engineered into the Col1A1 locus where its expression can be induced by Cre recombinase. Using organoids generated from this model, we evaluated its oncogenic activity and the biochemical effects of the fusion protein on the RHOA pathway and its downstream cell biological effects in the pathogenesis of DGC. RESULTS: We demonstrated that induction of CLDN18-ARHGAP26 expression in gastric organoids induced the formation of signet ring cells, characteristic features of DGC and was able to cooperatively transform gastric cells when combined with the loss of the tumour suppressor geneTrp53. CLDN18-ARHGAP26 promotes the activation of RHOA and downstream effector signalling. Molecularly, the fusion promotes activation of the focal adhesion kinase (FAK) and induction of the YAP pathway. A combination of FAK and YAP/TEAD inhibition can significantly block tumour growth. CONCLUSION: These results indicate that the CLDN18-ARHGAP26 fusion is a gain-of-function DGC oncogene that leads to activation of RHOA and activation of FAK and YAP signalling. These results argue for further evaluation of emerging FAK and YAP-TEAD inhibitors for these deadly cancers.

Bladder MRI with deep learning-based reconstruction: a prospective evaluation of muscle invasiveness using VI-RADS.

PURPOSE: To investigate the influence of deep learning reconstruction (DLR) on bladder MRI, specifically examination time, image quality, and diagnostic performance of vesical imaging reporting and data system (VI-RADS) within a prospective clinical cohort. METHODS: Seventy participants with bladder cancer who underwent MRI between August 2022 and February 2023 with a protocol containing standard T2-weighted imaging (T2WIS), standard diffusion-weighted imaging (DWIS), fast T2WI with DLR (T2WIDL), and fast DWI with DLR (DWIDL) were enrolled in this prospective study. Imaging quality was evaluated by measuring signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and qualitative image quality scoring. Additionally, the apparent diffusion coefficient (ADC) of bladder lesions derived from DWIS and DWIDL was measured and VI-RADS scoring was performed. Paired t-test or paired Wilcoxon signed-rank test were performed to compare image quality score, SNR, CNR, and ADC between standard sequences and fast sequences with DLR. The diagnostic performance for VI-RADS was assessed using the area under the receiver operating characteristic curve (AUC). RESULTS: Compared to T2WIS and DWIS, T2WIDL and DWIDL reduced the acquisition time from 5:57 min to 3:13 min and showed significantly higher SNR, CNR, qualitative image quality score of overall image quality, image sharpness, and lesion conspicuity. There were no significant differences in ADC and AUC of VI-RADS between standard sequences and fast sequences with DLR. CONCLUSIONS: The application of DLR to T2WI and DWI reduced examination time and significantly improved image quality, maintaining ADC and the diagnostic performance of VI-RADS for evaluating muscle invasion in bladder cancer.

MRI evaluation of vesical imaging reporting and data system for bladder cancer after neoadjuvant chemotherapy.

BACKGROUND: The Vesical Imaging-Reporting and Data System (VI-RADS) has demonstrated effectiveness in predicting muscle invasion in bladder cancer before treatment. The urgent need currently is to evaluate the muscle invasion status after neoadjuvant chemotherapy (NAC) for bladder cancer. This study aims to ascertain the accuracy of VI-RADS in detecting muscle invasion post-NAC treatment and assess its diagnostic performance across readers with varying experience levels. METHODS: In this retrospective study, patients with muscle-invasive bladder cancer who underwent magnetic resonance imaging (MRI) after NAC from September 2015 to September 2018 were included. VI-RADS scores were independently assessed by five radiologists, consisting of three experienced in bladder MRI and two inexperienced radiologists. Comparison of VI-RADS scores was made with postoperative histopathological diagnosis. Receiver operating characteristic curve analysis (ROC) was used for evaluating diagnostic performance, calculating sensitivity, specificity, and area under ROC (AUC)). Interobserver agreement was assessed using the weighted kappa statistic. RESULTS: The final analysis included 46 patients (mean age: 61 years ± 9 [standard deviation]; age range: 39-70 years; 42 men). The pooled AUC for predicting muscle invasion was 0.945 (95% confidence interval (CI): 0.893-0.977) for experienced readers, and 0.910 (95% CI: 0.831-0.959) for inexperienced readers, and 0.932 (95% CI: 0.892-0.961) for all readers. At an optimal cut-off value ≥ 4, pooled sensitivity and specificity were 74.1% (range: 66.0-80.9%) and 94.1% (range: 88.6-97.7%) for experienced readers, and 63.9% (range: 59.6-68.1%) and 86.4% (range: 84.1-88.6%) for inexperienced readers. Interobserver agreement ranged from substantial to excellent between all readers (k = 0.79-0.92). CONCLUSIONS: VI-RADS accurately assesses muscle invasion in bladder cancer patients after NAC and exhibits good diagnostic performance across readers with different experience levels.

Validation of GPT-4 for clinical event classification: A comparative analysis with ICD codes and human reviewers.

BACKGROUND AND AIM: Effective clinical event classification is essential for clinical research and quality improvement. The validation of artificial intelligence (AI) models like Generative Pre-trained Transformer 4 (GPT-4) for this task and comparison with conventional methods remains unexplored. METHODS: We evaluated the performance of the GPT-4 model for classifying gastrointestinal (GI) bleeding episodes from 200 medical discharge summaries and compared the results with human review and an International Classification of Diseases (ICD) code-based system. The analysis included accuracy, sensitivity, and specificity evaluation, using ground truth determined by physician reviewers. RESULTS: GPT-4 exhibited an accuracy of 94.4% in identifying GI bleeding occurrences, outperforming ICD codes (accuracy 63.5%, P < 0.001). GPT-4's accuracy was either slightly lower or statistically similar to individual human reviewers (Reviewer 1: 98.5%, P < 0.001; Reviewer 2: 90.8%, P = 0.170). For location classification, GPT-4 achieved accuracies of 81.7% and 83.5% for confirmed and probable GI bleeding locations, respectively, with figures that were either slightly lower or comparable with those of human reviewers. GPT-4 was highly efficient, analyzing the dataset in 12.7 min at a cost of 21.2 USD, whereas human reviewers required 8-9 h each. CONCLUSION: Our study indicates GPT-4 offers a reliable, cost-efficient, and faster alternative to current clinical event classification methods, outperforming the conventional ICD coding system and performing comparably to individual expert human reviewers. Its implementation could facilitate more accurate and granular clinical research and quality audits. Future research should explore scalability, prompt and model tuning, and ethical implications of high-performance AI models in clinical data processing.

Development of an Interpretable Deep Learning Model for Pathological Tumor Response Assessment After Neoadjuvant Therapy.

BACKGROUND: Neoadjuvant therapy followed by surgery has become the standard of care for locally advanced esophageal squamous cell carcinoma (ESCC) and accurate pathological response assessment is critical to assess the therapeutic efficacy. However, it can be laborious and inconsistency between different observers may occur. Hence, we aim to develop an interpretable deep-learning model for efficient pathological response assessment following neoadjuvant therapy in ESCC. METHODS: This retrospective study analyzed 337 ESCC resection specimens from 2020-2021 at the Pudong-Branch (Cohort 1) and 114 from 2021-2022 at the Puxi-Branch (External Cohort 2) of Fudan University Shanghai Cancer Center. Whole slide images (WSIs) from these two cohorts were generated using different scanning machines to test the ability of the model in handling color variations. Four pathologists independently assessed the pathological response. The senior pathologists annotated tumor beds and residual tumor percentages on WSIs to determine consensus labels. Furthermore, 1850 image patches were randomly extracted from Cohort 1 WSIs and binarily classified for tumor viability. A deep-learning model employing knowledge distillation was developed to automatically classify positive patches for each WSI and estimate the viable residual tumor percentages. Spatial heatmaps were output for model explanations and visualizations. RESULTS: The approach achieved high concordance with pathologist consensus, with an R^2 of 0.8437, a RAcc_0.1 of 0.7586, a RAcc_0.3 of 0.9885, which were comparable to two senior pathologists (R^2 of 0.9202/0.9619, RAcc_0.1 of 8506/0.9425, RAcc_0.3 of 1.000/1.000) and surpassing two junior pathologists (R^2 of 0.5592/0.5474, RAcc_0.1 of 0.5287/0.5287, RAcc_0.3 of 0.9080/0.9310). Visualizations enabled the localization of residual viable tumor to augment microscopic assessment. CONCLUSION: This work illustrates deep learning's potential for assisting pathological response assessment. Spatial heatmaps and patch examples provide intuitive explanations of model predictions, engendering clinical trust and adoption (Code and data will be available at https://github.com/WinnieLaugh/ESCC_Percentage once the paper has been conditionally accepted). Integrating interpretable computational pathology could help enhance the efficiency and consistency of tumor response assessment and empower precise oncology treatment decisions.

Single-cell transcriptomics dissects the transcriptome alterations of hematopoietic stem cells in myelodysplastic neoplasms.

BACKGROUND: Myelodysplastic neoplasms (MDS) are myeloid neoplasms characterized by disordered differentiation of hematopoietic stem cells and a predisposition to acute myeloid leukemia (AML). The underline pathogenesis remains unclear. METHODS: In this study, the trajectory of differentiation and mechanisms of leukemic transformation were explored through bioinformatics analysis of single-cell RNA-Seq data from hematopoietic stem and progenitor cells (HSPCs) in MDS patients. RESULTS: Among the HSPC clusters, the proportion of common myeloid progenitor (CMP) was the main cell cluster in the patients with excess blasts (EB)/ secondary AML. Cell cycle analysis indicated the CMP of MDS patients were in an active proliferative state. The genes involved in the cell proliferation, such as MAML3 and PLCB1, were up-regulated in MDS CMP. Further validation analysis indicated that the expression levels of MAML3 and PLCB1 in patients with MDS-EB were significantly higher than those without EB. Patients with high expression of PLCB1 had a higher risk of transformation to AML. PLCB1 inhibitor can suppress proliferation, induce cell cycle arrest, and activate apoptosis of leukemic cells in vitro. CONCLUSION: This study revealed the transcriptomic change of HSPCs in MDS patients along the pseudotime and indicated that PLCB1 plays a key role in the transformation of MDS into leukemia.