UroAngel: a single-kidney function prediction system based on computed tomography urography using deep learning.

BACKGROUND: Accurate estimation of the glomerular filtration rate (GFR) is clinically crucial for determining the status of obstruction, developing treatment strategies, and predicting prognosis in obstructive nephropathy (ON). We aimed to develop a deep learning-based system, named UroAngel, for non-invasive and convenient prediction of single-kidney function level. METHODS: We retrospectively collected computed tomography urography (CTU) images and emission computed tomography diagnostic reports of 520 ON patients. A 3D U-Net model was used to segment the renal parenchyma, and a logistic regression multi-classification model was used to predict renal function level. We compared the predictive performance of UroAngel with the Modification of Diet in Renal Disease (MDRD), Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations, and two expert radiologists in an additional 40 ON patients to validate clinical effectiveness. RESULTS: UroAngel based on 3D U-Net convolutional neural network could segment the renal cortex accurately, with a Dice similarity coefficient of 0.861. Using the segmented renal cortex to predict renal function stage had high performance with an accuracy of 0.918, outperforming MDRD and CKD-EPI and two radiologists. CONCLUSIONS: We proposed an automated 3D U-Net-based analysis system for direct prediction of single-kidney function stage from CTU images. UroAngel could accurately predict single-kidney function in ON patients, providing a novel, reliable, convenient, and non-invasive method.

Non-traumatic osteonecrosis of the femoral head induced by steroid and alcohol exposure is associated with intestinal flora alterations and metabolomic profiles.

OBJECTIVE: Osteonecrosis of the femoral head (ONFH) is a severe disease that primarily affects the middle-aged population, imposing a significant economic and social burden. Recent research has linked the progression of non-traumatic osteonecrosis of the femoral head (NONFH) to the composition of the gut microbiota. Steroids and alcohol are considered major contributing factors. However, the relationship between NONFH caused by two etiologies and the microbiota remains unclear. In this study, we examined the gut microbiota and fecal metabolic phenotypes of two groups of patients, and analyzed potential differences in the pathogenic mechanisms from both the microbial and metabolic perspectives. METHODS: Utilizing fecal samples from 68 NONFH patients (32 steroid-induced, 36 alcohol-induced), high-throughput 16 S rDNA sequencing and liquid chromatography with tandem mass spectrometry (LC-MS/MS) metabolomics analyses were conducted. Univariate and multivariate analyses were applied to the omics data, employing linear discriminant analysis effect size to identify potential biomarkers. Additionally, functional annotation of differential metabolites and associated pathways was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Subsequently, Spearman correlation analysis was employed to assess the potential correlations between differential gut microbiota and metabolites. RESULTS: High-throughput 16 S rDNA sequencing revealed significant gut microbial differences. At the genus level, the alcohol group had higher Lactobacillus and Roseburia, while the steroid group had more Megasphaera and Akkermansia. LC-MS/MS metabolomic analysis indicates significant differences in fecal metabolites between steroid- and alcohol-induced ONFH patients. Alcohol-induced ONFH (AONFH) showed elevated levels of L-Lysine and Oxoglutaric acid, while steroid-induced ONFH(SONFH) had increased Gluconic acid and Phosphoric acid. KEGG annotation revealed 10 pathways with metabolite differences between AONFH and SONFH patients. Correlation analysis revealed the association between differential gut flora and differential metabolites. CONCLUSIONS: Our results suggest that hormones and alcohol can induce changes in the gut microbiota, leading to alterations in fecal metabolites. These changes, driven by different pathways, contribute to the progression of the disease. The study opens new research directions for understanding the pathogenic mechanisms of hormone- or alcohol-induced NONFH, suggesting that differentiated preventive and therapeutic approaches may be needed for NONFH caused by different triggers.

Construction of a necroptosis-related lncRNA signature for predicting prognosis and revealing the immune microenvironment in bladder cancer.

BACKGROUND: Bladder cancer (BCa) is a common malignancy in the urinary system. Necroptosis, a recently discovered form of programmed cell death, is closely associated with the development and progression of various types of tumors. Targeting necroptosis through anti-cancer strategies has shown potential as a therapy for cancer. We aimed to develop a necroptosis-related lncRNAs (NRlncRNAs) risk model that can predict the survival and tumor immunity of BCa patients. METHODS: We analyzed sequencing data obtained from the TCGA database, and applied least absolute shrinkage and selection operator (LASSO) and Cox regression analysis to identify crucial NRlncRNAs for building a risk model. Using the risk score, we categorized patients into high- and low-risk groups, and assessed the accuracy with the area under the receiver operating characteristic (AUROC) and Kaplan-Meier curves. We performed the RT-qPCR to detect the expression differences of the genes based on the risk model. RESULTS: We identified a total of 296 NRlncRNAs, and 6 of them were included in the prognostic model. The AUC values for 1-, 3-, and 5-year predictions were 0.675, 0.726 and 0.734, respectively. Our risk model demonstrated excellent predictive performance and served as an independent predictor with high predictive power. Additionally, we performed PCA, TMB, GSEA analyses, and evaluated immune cell infiltration, to reveal significant differences between the high- and low-risk groups in functional signaling pathways, immunological status, and mutation profiles. Finally, we assessed the chemotherapeutic response of several drugs. According to the RT-qPCR results, we found that four NRlncRNAs of the risk model were more highly expressed in BCa cell lines than human immortalized uroepithelial cell line and regulated the occurrence and progression of bladder cancer. CONCLUSION: We constructed a novel NRlncRNAs-associated risk model, which could predict the prognosis and immune response of BCa patients.

Increased risk of periprosthetic joint infection after traumatic injury in joint revision patients.

BACKGROUND: Periprosthetic joint infection (PJI) is a serious complication after total joint arthroplasty (TJA). Although some risk factors of PJI were well studied, the association between trauma and PJI remains unknown in revision patients. MATERIALS AND METHODS: Between 2015 and 2018, a total of 71 patients with trauma history before revisions (trauma cohort) were propensity score matched (PSM) at a ratio of 1 to 5 with a control cohort of revision patients without a history of trauma. Then, the cumulative incidence rate of PJI within 3 years after operation between the two groups was compared. The secondary endpoints were aseptic revisions within 3 postoperative years, complications up to 30 postoperative days, and readmission up to 90 days. During a minimal 3-year follow-up, the survival was comparatively analyzed between the trauma cohort and the control cohort. RESULTS: The cumulative incidence of PJI was 40.85% in patients with trauma history against 27.04% in the controls (P = 0.02). Correspondingly, the cumulative incidence of aseptic re-revisions was 12.68% in patients with trauma history compared with 5.07% in the control cohort (P = 0.028). Cox regression revealed that trauma history was a risk factor of PJI (HR, 1.533 [95%CI, (1.019,2.306)]; P = 0.04) and aseptic re-revisions (HR, 3.285 [95%CI, (1.790,6.028)]; P < 0.0001). CONCLUSIONS: Our study demonstrated that revision patients with trauma history carried a higher risk of PJI compared to those without trauma history. Moreover, after revisions, the trauma patients were still at higher risk for treatment failure due to PJI, periprosthetic joint fracture, and mechanical complications.

The significance of an immunohistochemical marker-based panel in assisting the diagnosis of parathyroid carcinoma.

PURPOSE: Parathyroid carcinoma (PC) is an endocrine malignancy with a poor prognosis. However, the diagnosis of PC is still a difficult problem. A model with immunohistochemical (IHC) staining of 5 biomarkers has been reported from limited samples for the differential diagnosis of PC. In the present study, a series of IHC markers was applied in relatively large samples to optimize the diagnostic model for PC. METHODS: In this study, 44 patients with PC, 6 patients with atypical parathyroid tumors and 57 patients with parathyroid adenomas were included. IHC staining for parafibromin, Ki-67, galectin-3, protein-encoding gene product 9.5 (PGP9.5), E-cadherin, and enhancer of zeste homolog 2 (EZH2) was performed on formalin-fixed, paraffin-embedded tissue samples. The effects of clinical characteristics, surgical procedure, and IHC staining results of tumor tissues on the diagnosis and prognosis of PC were evaluated retrospectively. RESULTS: A logistic regression model with IHC results of parafibromin, Ki-67, and E-cadherin was created to differentiate PC with an area under the curve of 0.843. Cox proportional hazards analysis showed that negative parafibromin staining (hazard ratio: 3.26, 95% confidence interval: 1.28-8.34, P = 0.013) was related to the recurrence of PC. CONCLUSION: An IHC panel of parafibromin, Ki-67 and E-cadherin may help to distinguish PC from parathyroid neoplasms. Among the 6 IHC markers and clinical features examined, the risk factor related to PC recurrence was parafibromin staining loss.

The E3 ubiquitin ligase TRIM39 modulates renal fibrosis induced by unilateral ureteral obstruction through regulating proteasomal degradation of PRDX3.

Renal fibrosis is considered to be the ultimate pathway for various chronic kidney disease, with a complex etiology and great therapeutic challenges. Tripartite motif-containing (TRIM) family proteins have been shown to be involved in fibrotic diseases, but whether TRIM39 plays a role in renal fibrosis remain unexplored. In this study, we investigated the role of TRIM39 in renal fibrosis and its molecular mechanism. TRIM39 expression was analyzed in patients' specimens, HK-2 cells and unilateral ureteral obstruction (UUO) mice were used for functional and mechanistic studies. We found an upregulated expression of TRIM39 in renal fibrosis human specimens and models. In addition, TRIM39 knockdown was found efficient for alleviating renal fibrosis in both UUO mice and HK-2 cells. Mechanistically, we demonstrated that TRIM39 interacted with PRDX3 directly and induced ubiquitination degradation of PRDX3 at K73 and K149 through the K48 chain, which resulted in ROS accumulation and increased inflammatory cytokine generation, and further aggravated renal fibrosis. It provided an emerging potential target for the therapies of renal fibrosis.

A Smartphone Application-Based Remote Rehabilitation System for Post-Total Knee Arthroplasty Rehabilitation: A Randomized Controlled Trial.

BACKGROUND: Remote rehabilitation after total knee arthroplasty has gradually gained popularity in recent years. This study aimed to determine whether smartphone application-based remote rehabilitation could outperform home-based rehabilitation and outpatient guidance in terms of 12-week outcomes following primary unilateral total knee arthroplasty. METHODS: Patients who underwent primary unilateral total knee arthroplasty were recruited and randomly divided into a telerehabilitation group and a control group. A total of 100 patients were examined, with 50 each assigned to the telerehabilitation and control groups. In the telerehabilitation group, a telerehabilitation application was installed on the smartphones of the participants to allow postdischarge guidance. The primary outcomes were knee range of motion (ROM) at 12 weeks postoperatively. Secondary outcomes included the Western Ontario and McMaster Universities Osteoarthritis Index, Knee Society Score, The Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36), Five Times Sit-to-Stand Test (5xSST), Single-Leg Stance Test (SLST), satisfaction, rehabilitation costs, complication rate, and 90-day readmission rate. All outcomes were collected at 2, 6, and 12 weeks after surgery. RESULTS: At 12 weeks postoperatively, the telerehabilitation patients significantly outperformed the controls in terms of knee ROM (124 ± 8.7 versus 119 ± 5.5 P = .01), SF-36 (physiological function) (61.5 ± 20.3 versus 45.5 ± 18.1 P = .000), SF-36 (role-physical) (49.3 ± 41.5 versus 27.7 ± 28.9 P = .012), SLST (13.0 ± 9.1 versus 9.1 ± 5.9 P = .026), and 5xSST (17.7 ± 4.3 versus 19.4 ± 3.5 P = .043). No significant differences were found between groups in the Western Ontario and McMaster Universities Osteoarthritis Index score, Knee Society Score, rehabilitation costs, 90-day readmission rate, or incidence of adverse events. CONCLUSION: Our study showed that smartphone app-based remote rehabilitation worked better than home-based rehabilitation with outpatient guidance in terms of short-term results in ROM, SLST, and 5xSST.

Overcoming the compensatory increase in NRF2 induced by NPL4 inhibition enhances disulfiram/copper-induced oxidative stress and ferroptosis in renal cell carcinoma.

Renal cell carcinoma (RCC) is the most common type of kidney cancer, and it appears to be highly susceptible to ferroptosis. Disulfiram, an alcoholism drug, has been shown to have anticancer properties in various studies, including those on RCC. However, the mechanism of the anticancer effect of disulfiram/copper on RCC remains unclear. In this study, we investigated the impact of disulfiram/copper on RCC treatment using both RCC cells and mouse subcutaneous tumor models. Our findings demonstrate that disulfiram/copper treatment reduced the viability of RCC cells, inhibited their invasion and migration, and disrupted mitochondrial homeostasis, ultimately leading to oxidative stress and ferroptosis. Mechanistically, disulfiram/copper treatment prolonged the half-life of NRF2 and reduced its degradation, but had no effect on transcription, indicating that the disulfiram/copper-induced increase in NRF2 was not related to transcription. Furthermore, we observed that disulfiram/copper treatment reduced the expression of NPL4, a ubiquitin protein-proteasome system involved in NRF2 degradation, while overexpression of NPL4 reversed NRF2 levels and enhanced disulfiram/copper-induced oxidative stress and ferroptosis. These results suggest that overcoming the compensatory increase in NRF2 induced by NPL4 inhibition enhances disulfiram/copper-induced oxidative stress and ferroptosis in RCC. In addition, our in vivo experiments revealed that disulfiram/copper synergized with sorafenib to inhibit the growth of RCC cells and induce ferroptosis. In conclusion, our study sheds light on a possible mechanism for disulfiram/copper treatment in RCC and provides a potential synergistic strategy to overcome sorafenib resistance.

Puerarin alleviated oxidative stress and ferroptosis during renal fibrosis induced by ischemia/reperfusion injury via TLR4/Nox4 pathway in rats.

PURPOSE: To investigate the role of puerarin on renal fibrosis and the underlying mechanism in renal ischemia and reperfusion (I/R) model. METHODS: Rats were intraperitoneally injected with puerarin (50 or 100 mg/kg) per day for one week before renal I/R. The level of renal collagen deposition and interstitial fibrosis were observed by hematoxylin and eosin and Sirius Red staining, and the expression of α-smooth muscle actin (α-SMA) was examined by immunohistochemical staining. The ferroptosis related factors and TLR4/Nox4-pathway-associated proteins were detected by Western blotting. RESULTS: Puerarin was observed to alleviate renal collagen deposition, interstitial fibrosis and the α-SMA expression induced by I/R. Superoxide dismutase (SOD) activities and glutathione (GSH) level were decreased in I/R and hypoxia/reoxygenation (H/R), whereas malondialdehyde (MDA) and Fe2+ level increased. However, puerarin reversed SOD, MDA, GSH and Fe2+ level changes induced by I/R and H/R. Besides, Western blot indicated that puerarin inhibited the expression of ferroptosis related factors in a dose-dependent manner, which further demonstrated that puerarin had the effect to attenuate ferroptosis. Moreover, the increased expression of TLR/Nox4-pathway-associated proteins were observed in I/R and H/R group, but puerarin alleviated the elevated TLR/Nox4 expression. CONCLUSIONS: Our results suggested that puerarin inhibited oxidative stress and ferroptosis induced by I/R and, thus, delayed the progression of renal fibrosis, providing a new target for the treatment of renal fibrosis.

A Weakly Supervised Deep Learning Model and Human-Machine Fusion for Accurate Grading of Renal Cell Carcinoma from Histopathology Slides.

(1) Background: The Fuhrman grading (FG) system is widely used in the management of clear cell renal cell carcinoma (ccRCC). However, it is affected by observer variability and irreproducibility in clinical practice. We aimed to use a deep learning multi-class model called SSL-CLAM to assist in diagnosing the FG status of ccRCC patients using digitized whole slide images (WSIs). (2) Methods: We recruited 504 eligible ccRCC patients from The Cancer Genome Atlas (TCGA) cohort and obtained 708 hematoxylin and eosin-stained WSIs for the development and internal validation of the SSL-CLAM model. Additionally, we obtained 445 WSIs from 188 ccRCC eligible patients in the Clinical Proteomic Tumor Analysis Consortium (CPTAC) cohort as an independent external validation set. A human-machine fusion approach was used to validate the added value of the SSL-CLAM model for pathologists. (3) Results: The SSL-CLAM model successfully diagnosed the five FG statuses (Grade-0, 1, 2, 3, and 4) of ccRCC, and achieved AUCs of 0.917 and 0.887 on the internal and external validation sets, respectively, outperforming a junior pathologist. For the normal/tumor classification (Grade-0, Grade-1/2/3/4) task, the SSL-CLAM model yielded AUCs close to 1 on both the internal and external validation sets. The SSL-CLAM model achieved a better performance for the two-tiered FG (Grade-0, Grade-1/2, and Grade-3/4) task, with AUCs of 0.936 and 0.915 on the internal and external validation sets, respectively. The human-machine diagnostic performance was superior to that of the SSL-CLAM model, showing promising prospects. In addition, the high-attention regions of the SSL-CLAM model showed that with an increasing FG status, the cell nuclei in the tumor region become larger, with irregular contours and increased cellular pleomorphism. (4) Conclusions: Our findings support the feasibility of using deep learning and human-machine fusion methods for FG classification on WSIs from ccRCC patients, which may assist pathologists in making diagnostic decisions.

Predicting Lymph Node Metastasis Status from Primary Muscle-Invasive Bladder Cancer Histology Slides Using Deep Learning: A Retrospective Multicenter Study.

BACKGROUND: Accurate prediction of lymph node metastasis (LNM) status in patients with muscle-invasive bladder cancer (MIBC) before radical cystectomy can guide the use of neoadjuvant chemotherapy and the extent of pelvic lymph node dissection. We aimed to develop and validate a weakly-supervised deep learning model to predict LNM status from digitized histopathological slides in MIBC. METHODS: We trained a multiple instance learning model with an attention mechanism (namely SBLNP) from a cohort of 323 patients in the TCGA cohort. In parallel, we collected corresponding clinical information to construct a logistic regression model. Subsequently, the score predicted by the SBLNP was incorporated into the logistic regression model. In total, 417 WSIs from 139 patients in the RHWU cohort and 230 WSIs from 78 patients in the PHHC cohort were used as independent external validation sets. RESULTS: In the TCGA cohort, the SBLNP achieved an AUROC of 0.811 (95% confidence interval [CI], 0.771-0.855), the clinical classifier achieved an AUROC of 0.697 (95% CI, 0.661-0.728) and the combined classifier yielded an improvement to 0.864 (95% CI, 0.827-0.906). Encouragingly, the SBLNP still maintained high performance in the RHWU cohort and PHHC cohort, with an AUROC of 0.762 (95% CI, 0.725-0.801) and 0.746 (95% CI, 0.687-0.799), respectively. Moreover, the interpretability of SBLNP identified stroma with lymphocytic inflammation as a key feature of predicting LNM presence. CONCLUSIONS: Our proposed weakly-supervised deep learning model can predict the LNM status of MIBC patients from routine WSIs, demonstrating decent generalization performance and holding promise for clinical implementation.

Long-term outcomes of arthroscopic synovectomy and core decompression through multiple small bone holes for early-stage avascular necrosis of the femoral head.

OBJECTIVE: This study described a minimally invasive approach for the management of early-stage avascular necrosis of the femoral head, which integrated arthroscopic intra-articular decompression and core decompression by drilling multiple small holes. METHOD: A total of 126 patients with 185 hip avascular necrosis were included between March 2005 and January 2008, and the hips were classified, based on the Association Research Circulation Osseous staging system, into stage I (n = 43), stage II (n = 114), and stage III (n = 28). Arthroscopic intra-articular inspection and debridement, along with drilling of multiple small holes for core decompression, were performed. The Modified Harris hip score system and radiographs were used to assess the pre- and post-surgery outcomes. RESULTS: One hundred and three patients (involving 153 hips) were followed up successfully for an average of 10.7 ± 3.4 years (range: 9-12 years). After surgery, the overall survival rate was 51.6% (79 hips), and the clinical survival rates were 79%, 72%, 52%, 32%, and 10% for patients with stage I, IIa, IIb, IIc, and III, respectively. The outcomes of patients with Association Research Circulation Osseous Stages I or IIA were better than those of other stages, while hips with a large necrotic area had poor results. This approach preserved the original biomechanical strength of the femoral head after core decompression and eliminated arthritis factors in the hip joint. CONCLUSION: The core decompression with multiple small-size holes is an effective method for treating early-stage avascular necrosis of the femoral head, particularly in those with pathological changes in the hip joint. LEVEL OF EVIDENCE: Therapeutic study, Level IV.

Machine Learning Quantified Tumor-Stroma Ratio Is an Independent Prognosticator in Muscle-Invasive Bladder Cancer.

Although the tumor-stroma ratio (TSR) has prognostic value in many cancers, the traditional semi-quantitative visual assessment method has inter-observer variability, making it impossible for clinical practice. We aimed to develop a machine learning (ML) algorithm for accurately quantifying TSR in hematoxylin-and-eosin (H&E)-stained whole slide images (WSI) and further investigate its prognostic effect in patients with muscle-invasive bladder cancer (MIBC). We used an optimal cell classifier previously built based on QuPath open-source software and ML algorithm for quantitative calculation of TSR. We retrospectively analyzed data from two independent cohorts to verify the prognostic significance of ML-based TSR in MIBC patients. WSIs from 133 MIBC patients were used as the discovery set to identify the optimal association of TSR with patient survival outcomes. Furthermore, we performed validation in an independent external cohort consisting of 261 MIBC patients. We demonstrated a significant prognostic association of ML-based TSR with survival outcomes in MIBC patients (p < 0.001 for all comparisons), with higher TSR associated with better prognosis. Uni- and multivariate Cox regression analyses showed that TSR was independently associated with overall survival (p < 0.001 for all analyses) after adjusting for clinicopathological factors including age, gender, and pathologic stage. TSR was found to be a strong prognostic factor that was not redundant with the existing staging system in different subgroup analyses (p < 0.05 for all analyses). Finally, the expression of six genes (DACH1, DEEND2A, NOTCH4, DTWD1, TAF6L, and MARCHF5) were significantly associated with TSR, revealing possible potential biological relevance. In conclusion, we developed an ML algorithm based on WSIs of MIBC patients to accurately quantify TSR and demonstrated its prognostic validity for MIBC patients in two independent cohorts. This objective quantitative method allows application in clinical practice while reducing the workload of pathologists. Thus, it might be of significant aid in promoting precise pathology services in MIBC.

Leukocyte Esterase Strip Quantitative Detection Based on RGB Photometry is a Probable Method to Diagnose Periprosthetic Joint Infection: An Exploratory Study.

OBJECTIVE: Leucocyte esterase (LE) strip test is the most rapid, convenient, and cheap method to diagnose chronic periprosthesis joint infection (PJI). However, the determination of LE strip mainly relies on colorimetric method with strong subjectivity, which leads to low diagnostic accuracy. Therefore, we try to convert LE strip images into digital data through the RGB photometric system to achieve objective diagnosis. This method will greatly improve the accuracy of LE strip detection and diagnosis of PJI. METHODS: From January 2021 to September 2021, 46 patients with suspected PJI after total hip and knee arthroplasty underwent diagnostic joint puncture. After effective joint fluid samples were harvested, they were divided into original fluid and centrifuged fluid for LE strip detection. Real-time images of LE strip were taken at 90 s, 3 min, 5 min, 10 min, and 15 min after sampling, and their brightness (Y) was obtained after they were input into an RGB photometric system. Grouping was based on centrifugation, infection, and time points, and then the differences in brightness among groups were compared. The correlation between LE strip image brightness and WBC count was evaluated. Student t-test was used for the parametric data and chi-square test for qualitative data. Simple linear regression was utilized to analyze the correlation between brightness and WBC count in each group. RESULTS: Included were 19 cases of PJI and 27 Non-PJI subjects diagnosed against ICM2018 diagnostic criteria. The brightness was lower in the PJI group than in Non-PJI group (p < 0.05). The brightness of the uncentrifuged group was lower than that of the centrifuged group (p < 0.05). Irrespective of centrifugation or infection, the brightness of LE strip decreased with the exposure time after sampling. The brightness of LE strip was correlated with WBC count at different time points, with the correlation being strongest 5 min after sampling (R2 (5 min) = 0.86, p < 0.0001). The correlation between LE strip brightness and WBC count was also found in the centrifugation group, with the correlation being most robust 15 min after sampling (R2 (15 min) = 0.73, p < 0.0001). CONCLUSION: A remarkable correlation was found between LE strip brightness and the WBC count. It is feasible to directly quantify LE strip image on a RGB photometer to achieve quantitative detection of LE strip to diagnose PJI.

Breast cancer heterogeneity and its implication in personalized precision therapy.

Breast cancer heterogeneity determines cancer progression, treatment effects, and prognosis. However, the precise mechanism for this heterogeneity remains unknown owing to its complexity. Here, we summarize the origins of breast cancer heterogeneity and its influence on disease progression, recurrence, and therapeutic resistance. We review the possible mechanisms of heterogeneity and the research methods used to analyze it. We also highlight the importance of cell interactions for the origins of breast cancer heterogeneity, which can be further categorized into cooperative and competitive interactions. Finally, we provide new insights into precise individual treatments based on heterogeneity.

Development and validation of a deep learning-based laparoscopic system for improving video quality.

PURPOSE: A clear surgical field of view is a prerequisite for successful laparoscopic surgery. Surgical smoke, image blur, and lens fogging can affect the clarity of laparoscopic imaging. We aimed to develop a real-time assistance system (namely LVQIS) for removing these interfering factors during laparoscopic surgery, thereby improving laparoscopic video quality. METHODS: LVQIS was developed with generative adversarial networks (GAN) and transfer learning, which included two classification models (ResNet-50), a motion blur removal model (MPRNet), and a smoke/fog removal model (GAN). 136 laparoscopic surgery videos were retrospectively collected in a tripartite dataset for training and validation. A synthetic dataset was simulated using the image enhancement library Albumentations and the 3D rendering software Blender. The objective evaluation results were through PSNR, SSIM and FID, and the subjective evaluation includes the operation pause time and the degree of anxiety of surgeons. RESULTS: The synthesized dataset contained 19,245 clear images, 19,245 motion blur images, and 19,245 smoke/fog images. The ResNet-50 CNN model identified whether a single laparoscopic image had motion blur and smoke/fog with an accuracy of over 0.99. The PSNR, SSIM and FID of the de-smoke model were 29.67, 0.9551 and 74.72, respectively, and the PSNR, SSIM and FID of the de-blurring model were 26.78, 0.9020 and 80.10, respectively, which were better than other advanced de-blurring and de-smoke/fog models. In a comparative study of 100 laparoscopic surgeries, the use of LVQIS significantly reduced the operation pause time (P < 0.001) and the anxiety of surgeons (P = 0.004). CONCLUSIONS: In this study, LVQIS is an efficient and robust system that can improve the quality of laparoscopic video, reduce surgical pause time and the anxiety of surgeons, and has the potential for real-time application in real clinical settings.

Development and validation of a deep-learning based assistance system for enhancing laparoscopic control level.

BACKGROUND AND AIMS: Inter-operator variations in the level of intraoperative laparoscope control by surgeons influence surgical outcomes. We aimed to construct a laparoscopic surgery quantification system (LSQS) for real-time evaluation of the surgeon's laparoscope control to improve intraoperative manipulation of the laparoscope. METHODS: Using 1888 images from 80 laparoscopic videos for training, the U-Net, PSPNet, LinkNet, and DeepLabv3+ models were used to segment surgical instruments. The percentage of the instruments in central area was defined as the new indicator and the threshold was determined from 20 laparoscopic videos. The differences between expert and non-expert laparoscopic operators before and after LSQS were compared. RESULTS: Among the three segmentation models (U-Net, PSPNet, and LinkNet), the PSPNet model had the highest index (precision 0.9135; F1 score 0.9058; mIoU 0.8280). The validation experiment showed that LSQS could help non-expert users to more easily achieve expert-level control of the laparoscope. CONCLUSIONS: Deep-learning technology successfully fed back real-time intraoperative information on level of laparoscope control and may facilitate better visualisation of the surgical field.

Different uptake patterns of 68Ga-FAPI in aseptic loosening and periprosthetic joint infection of hip arthroplasty: A case series and literature review.

Background: The diagnosis of a periprosthetic joint infection (PJI) is always a difficult point in research on the surgery of joints. The current diagnostic criteria include a comprehensive analysis of multiple tests; however, there are no effective visual examinations yet that can differentiate between aseptic loosening and the PJI. Case presentation: This case report describes four patients with symptomatic total hip arthroplasty (THA), two cases of loosening and two cases of infection. Although the four cases were correctly diagnosed by the tissue culture, preoperative tests and pathological examination could not effectively distinguish an infection from a non-infection. Based on a preclinical study and theoretical feasibility, gallium-68 (68Ga)-labeled fibroblast activation protein inhibitor positron emission tomography/computed tomography (68Ga-FAPI PET/CT) was performed. Through 68Ga-FAPI PET/CT scanning, not only were the causes diagnosed correctly but the lesions were also located. Conclusion: When the lesion is located between the bone and the prosthesis, 68Ga-FAPI PET/CT could differentiate aseptic loosening from periprosthetic joint infection (PJI). 68Ga-FAPI PET/CT has clear advantages over routine examinations and has a prospective application in detecting PJI.

Accurate Diagnosis and Survival Prediction of Bladder Cancer Using Deep Learning on Histological Slides.

(1) Background: Early diagnosis and treatment are essential to reduce the mortality rate of bladder cancer (BLCA). We aimed to develop deep learning (DL)-based weakly supervised models for the diagnosis of BLCA and prediction of overall survival (OS) in muscle-invasive bladder cancer (MIBC) patients using whole slide digitized histological images (WSIs). (2) Methods: Diagnostic and prognostic models were developed using 926 WSIs of 412 BLCA patients from The Cancer Genome Atlas cohort. We collected 250 WSIs of 150 BLCA patients from the Renmin Hospital of Wuhan University cohort for external validation of the models. Two DL models were developed: a BLCA diagnostic model (named BlcaMIL) and an MIBC prognostic model (named MibcMLP). (3) Results: The BlcaMIL model identified BLCA with accuracy 0.987 in the external validation set, comparable to that of expert uropathologists and outperforming a junior pathologist. The C-index values for the MibcMLP model on the internal and external validation sets were 0.631 and 0.622, respectively. The risk score predicted by MibcMLP was a strong predictor independent of existing clinical or histopathologic indicators, as demonstrated by univariate Cox (HR = 2.390, p < 0.0001) and multivariate Cox (HR = 2.414, p < 0.0001) analyses. The interpretability of DL models can help in the analysis of critical regions associated with tumors to enrich the information obtained from WSIs. Furthermore, the expression of six genes (ANAPC7, MAPKAPK5, COX19, LINC01106, AL161431.1 and MYO16-AS1) was significantly associated with MibcMLP-predicted risk scores, revealing possible potential biological correlations. (4) Conclusions: Our study developed DL models for accurately diagnosing BLCA and predicting OS in MIBC patients, which will help promote the precise pathological diagnosis of BLCA and risk stratification of MIBC to improve clinical treatment decisions.