One-class classification aims to learn one-class models from only in-class training samples. Because of lacking out-of-class samples during training, most conventional deep learning based methods ...suffer from the feature collapse problem. In contrast, contrastive learning based methods can learn features from only in-class samples but are hard to be end-to-end trained with one-class models. To address the aforementioned problems, we propose alternating direction method of multipliers based sparse representation network (ADMM-SRNet). ADMM-SRNet contains the heterogeneous contrastive feature (HCF) network and the sparse dictionary (SD) network. The HCF network learns in-class heterogeneous contrastive features by using contrastive learning with heterogeneous augmentations. Then, the SD network models the distributions of the in-class training samples by using dictionaries computed based on ADMM. By coupling the HCF network, SD network and the proposed loss functions, our method can effectively learn discriminative features and one-class models of the in-class training samples in an end-to-end trainable manner. Experimental results show that the proposed method outperforms state-of-the-art methods on CIFAR-10, CIFAR-100 and ImageNet-30 datasets under one-class classification settings.
Objective
Early diagnosis and treatment of nasopharyngeal carcinoma (NPC) are vital for a better prognosis. Still, because of obscure anatomical sites and insidious symptoms, nearly 80% of patients ...with NPC are diagnosed at a late stage. This study aimed to validate a machine learning (ML) model utilizing symptom‐related diagnoses and procedures in medical records to predict nasopharyngeal carcinoma (NPC) occurrence and reduce the prediagnostic period.
Materials and Methods
Data from a population‐based health insurance database (2001–2008) were analyzed, comparing adults with and without newly diagnosed NPC. Medical records from 90 to 360 days before diagnosis were examined. Five ML algorithms (Light Gradient Boosting Machine LGB, eXtreme Gradient Boosting XGB, Multivariate Adaptive Regression Splines MARS, Random Forest RF, and Logistics Regression LG) were evaluated for optimal early NPC detection. We further use a real‐world data of 1 million individuals randomly selected for testing the final model. Model performance was assessed using AUROC. Shapley values identified significant contributing variables.
Results
LGB showed maximum predictive power using 14 features and 90 days before diagnosis. The LGB models achieved AUROC, specificity, and sensitivity were 0.83, 0.81, and 0.64 for the test dataset, respectively. The LGB‐driven NPC predictive tool effectively differentiated patients into high‐risk and low‐risk groups (hazard ratio: 5.85; 95% CI: 4.75–7.21). The model‐layering effect is valid.
Conclusions
ML approaches using electronic medical records accurately predicted NPC occurrence. The risk prediction model serves as a low‐cost digital screening tool, offering rapid medical decision support to shorten prediagnostic periods. Timely referral is crucial for high‐risk patients identified by the model.
The LGB‐driven NPC predictive model showed optimal predictive power (AUROC: 0.83), and effectively differentiated patients into high‐risk and low‐risk groups (HR: 5.85). It serve as a low‐cost digital screening tool, offering rapid medical decision support to shorten prediagnostic periods. Timely referral is crucial for high‐risk patients identified by the model.
Oral anticoagulants (OAC) are recommended for preventing stroke and systemic embolism in atrial fibrillation. Proper use is imperative for maximizing anticoagulation therapy’s effectiveness and ...safety. In preparation for the implementation of a smartphone-based SmartMed app (application) aiming to promote patient self-management, medication adherence, and data collection for patients on anticoagulation therapy, its usability assessment can ensure the value of OAC app development and adoption. We evaluated the SmartMed app’s usability using the System Usability Scale (SUS) and the app-specific domain of the Mobile App Rating Scale (MARS) for its perceived impact on taking OAC regularly. We recruited 25 OAC users and their home caregivers and 59 healthcare professionals, including pharmacists, nurses, and cardiac surgeons from one medical center and one regional hospital in Taiwan. All participants (n = 84) thought the SmartMed app was useful, with mean SUS and MARS scores of 81.49 (±14.42) and 4.65 (±0.49), respectively. Usability evaluation revealed that fewer experiences with smartphone apps and different healthcare professionals (pharmacists versus nurses or cardiac surgeons) were associated with lower SUS scores and perceived impact. Throughout the evaluation process, the SmartMed app’s design was considered helpful from multiple stakeholders’ perspectives. Further ongoing mobile technology supports are necessary to establish the SmartMed app’s effectiveness.
Background: Long-term care (LTC) service demands among cancer patients are significantly understudied, leading to gaps in healthcare resource allocation and policymaking. Objective: This study aimed ...to predict LTC service demands for cancer patients and identify the crucial factors. Methods: 3333 cases of cancers were included. We further developed two specialized prediction models: a Unified Prediction Model (UPM) and a Category-Specific Prediction Model (CSPM). The UPM offered generalized forecasts by treating all services as identical, while the CSPM built individual predictive models for each specific service type. Sensitivity analysis was also conducted to find optimal usage cutoff points for determining the usage and non-usage cases. Results: Service usage differences in lung, liver, brain, and pancreatic cancers were significant. For the UPM, the top 20 performance model cutoff points were adopted, such as through Logistic Regression (LR), Quadratic Discriminant Analysis (QDA), and XGBoost (XGB), achieving an AUROC range of 0.707 to 0.728. The CSPM demonstrated performance with an AUROC ranging from 0.777 to 0.837 for the top five most frequently used services. The most critical predictive factors were the types of cancer, patients’ age and female caregivers, and specific health needs. Conclusion: The results of our study provide valuable information for healthcare decisions, resource allocation optimization, and personalized long-term care usage for cancer patients.
Background & Aims The incidence of hepatocellular carcinoma (HCC) increases with age, but protective antibody responses decrease with time after infants are immunized against hepatitis B virus (HBV). ...We investigated whether immunization of infants against HBV prevents their developing HCC as adults. We also searched for strategies to maximize the cancer-preventive effects. Methods We collected data from 2 Taiwan HCC registry systems on 1509 patients (6–26 years old) diagnosed with HCC from 1983 through 2011. Data on history of HBV immunization and prenatal maternal levels of HBV antigens of all HCC patients born after July 1984 were retrieved from the HBV immunization data bank of the Taiwan Center for Disease Control. We collected data on birth cohort-specific populations (6–26 years old) of Taiwan using the National Household Registry System. Rates of HCC incidence per 105 person-years were derived by dividing the number of patients with HCC by the person-years of the general population. Relative risks (RR) for HCC were estimated by Poisson regression analysis in vaccinated vs unvaccinated birth cohorts. We stratified patients by age group to evaluate the association of birth cohorts and HCC risks. Results Of the 1509 patients with HCC, 1343 were born before, and 166 were born after, the HBV vaccination program began. HCC incidence per 105 person-years was 0.92 in the unvaccinated cohort and 0.23 in the vaccinated birth cohorts. The RRs for HCC in patients 6–9 years old, 10–14 years old, 15–19 years old, and 20–26 years old who were vaccinated vs unvaccinated were 0.26 (95% confidence interval CI, 0.17–0.40), 0.34 (95% CI, 0.25–0.48), 0.37 (95% CI, 0.25–0.51), and 0.42 (95% CI, 0.32–0.56), respectively. The RR for HCC in 6- to 26-year-olds was lower in the later vs the earlier cohorts (born in 1992–2005 vs 1986–1992; P < .001 and 1986–1992 vs 1984–1986; P < .002). Transmission of HBV from highly infectious mothers and incomplete immunization were associated with development of HCC. Conclusions Based on an analysis of 1509 patients with HCC in Taiwan, immunization of infants against HBV reduces their risk of developing HCC as children and young adults. Improving HBV vaccination strategies and overcoming risk factors could reduce the incidence of liver cancer.
We collected 110 Salmonella enterica isolates from sick pigs and determined their serotypes, genotypes using pulsed-field gel electrophoresis (PFGE), and antimicrobial susceptibility to 12 ...antimicrobials and compared the data with a collection of 18,280 isolates obtained from humans. The pig isolates fell into 12 common serovars for human salmonellosis in Taiwan; S. Typhimurium, S. Choleraesuis, S. Derby, S. Livingstone, and S. Schwarzengrund were the 5 most common serovars and accounted for a total of 84% of the collection. Of the 110 isolates, 106 (96%) were multidrug resistant (MDR) and 48 (44%) had PFGE patterns found in human isolates. S. Typhimurium, S. Choleraesuis, and S. Schwarzengrund were among the most highly resistant serovars. The majority of the 3 serovars were resistant to 8-11 of the tested antimicrobials. The isolates from pigs and humans sharing a common PFGE pattern displayed identical or very similar resistance patterns and Salmonella strains that caused severe infection in pigs were also capable of causing infections in humans. The results indicate that pigs are one of the major reservoirs to human salmonellosis in Taiwan. Almost all of the pig isolates were MDR, which highlights the necessity of strictly regulating the use of antimicrobials in the agriculture sector in Taiwan.
Oral squamous cell carcinoma (OSCC), like many solid tumors, contains a heterogeneous population of cancer cells. Recent data suggest that a rare subpopulation of cancer cells, termed cancer stem ...cells (CSC), is capable of initiating, maintaining, and expanding the growth of tumor. Identification and characterization of CSC from OSCC facilitates the monitoring, therapy, or prevention of OSCC.
We enriched oral cancer stem-like cells (OC-SLC) through sphere formation by cultivating OSCC cells from established OSCC cell lines or primary cultures of OSCC patients within defined serum-free medium. Differential expression profile of stemness genes between enriched OC-SLC and parental OSCC was elucidated. Furthermore, immunohistochemical staining of stemness markers on OSCC patient tissues was examined to evaluate the association between stemness genes and prognosis of OSCC.
Enriched OC-SLC highly expressed the stem/progenitor cell markers and ABC transporter gene (Oct-4, Nanog, CD117, Nestin, CD133, and ABCG2) and also displayed induced differentiation abilities and enhanced migration/invasion/malignancy capabilities in vitro and in vivo. Elevated expression of CD133 was shown in the enriched OC-SLC from OSCC patients' tumors. Positive correlations of Oct-4, Nanog, or CD133 expression on tumor stage were shown on 52 OSCC patient tissues. Kaplan-Meier analyses exhibited that Nanog/Oct-4/CD133 triple-positive patients predicted the worst survival prognosis of OSCC patients.
We enriched a subpopulation of cancer stem-like cell from OSCC by sphere formation. The enriched OC-SLC possesses the characteristics of both stem cells and malignant tumors. Additionally, expression of stemness markers (Nanog/Oct-4/CD133) contradicts the survival prognosis of OSCC patients.
Most cases of oral squamous cell carcinoma (OSCC) develop from visible oral potentially malignant disorders (OPMDs). The latter exhibit heterogeneous subtypes with different transformation ...potentials, complicating the early detection of OSCC during routine visual oral cancer screenings. To develop clinically applicable biomarkers, we collected saliva samples from 96 healthy controls, 103 low-risk OPMDs, 130 high-risk OPMDs, and 131 OSCC subjects. These individuals were enrolled in Taiwan’s Oral Cancer Screening Program. We identified 302 protein biomarkers reported in the literature and/or through in-house studies and prioritized 49 proteins for quantification in the saliva samples using multiple reaction monitoring-MS. Twenty-eight proteins were successfully quantified with high confidence. The quantification data from non-OSCC subjects (healthy controls + low-risk OPMDs) and OSCC subjects in the training set were subjected to classification and regression tree analyses, through which we generated a four-protein panel consisting of MMP1, KNG1, ANXA2, and HSPA5. A risk-score scheme was established, and the panel showed high sensitivity (87.5%) and specificity (80.5%) in the test set to distinguish OSCC samples from non-OSCC samples. The risk score >0.4 detected 84% (42/50) of the stage I OSCCs and a significant portion (42%) of the high-risk OPMDs. Moreover, among 88 high-risk OPMD patients with available follow-up results, 18 developed OSCC within 5 y; of them, 77.8% (14/18) had risk scores >0.4. Our four-protein panel may therefore offer a clinically effective tool for detecting OSCC and monitoring high-risk OPMDs through a readily available biofluid.
Prostate cancer (PCa), an extremely common malignancy in males, is the most prevalent disease in several countries. Norcantharidin (NCTD) has antiproliferation, antimetastasis, apoptosis, and ...autophagy effects in various tumor cells. Nevertheless, the antitumor effect of NCTD combined with paclitaxel (PTX), a chemotherapeutic drug, in PCa remains unknown. The cell growth, proliferative rate, cell cycle distribution, and cell death were determined by 3‐4,5‐dimethylthiazol‐2‐yl‐2,5 diphenyltetrazolium bromide, colony formation assay, PI staining, and Annexin V/PI staining by flow cytomertry, whereas the mitochondrial membrane potential (MMP) and endoplasmic reticulum (ER) stress was evaluated using the MitoPotential assay and ER‐ID red assay. We also evaluated the protein and mRNA expression of SIRTs by Western blotting and qRTPCR assay. Overexpression effectivity was measured by DNA transfection assay. Our study showed that cell viability and proliferative PC3 and DU145 rates were effectively inhibited after NCTD‐PTX combination. We also found that NCTD‐PTX combination treatment significantly enhance G2/M phase arrest, induction of cell death and ER stress, loss of MMP, and ER‐ or apoptotic‐related protein expression. Furthermore, NCTD‐PTX combination treatment was significantly decreasing the protein and mRNA expression of SIRT7 in PCa cells. Combination therapy effectively reduced cell viability, ER stress‐mediated apoptosis and p‐eIF2α/ATF4/CHOP/cleaved‐PARP expression inhibition in SIRT7 overexpression of PCa cells. These results indicate that NCTD combined with PTX induces ER stress‐mediated apoptosis of PCa cells by regulating the SIRT7 expression axis. Moreover, combination therapy may become a potential therapeutic strategy against human PCa.
The efficient and safe delivery of therapeutic drugs, proteins, and nucleic acids are essential for meaningful therapeutic benefits. The field of nanomedicine shows promising implications in the ...development of therapeutics by delivering diagnostic and therapeutic compounds. Nanomedicine development has led to significant advances in the design and engineering of nanocarrier systems with supra-molecular structures. Smart mesoporous silica nanoparticles (MSNs), with excellent biocompatibility, tunable physicochemical properties, and site-specific functionalization, offer efficient and high loading capacity as well as robust and targeted delivery of a variety of payloads in a controlled fashion. Such unique nanocarriers should have great potential for challenging biomedical applications, such as tissue engineering, bioimaging techniques, stem cell research, and cancer therapies. However, in vivo applications of these nanocarriers should be further validated before clinical translation. To this end, this review begins with a brief introduction of MSNs properties, targeted drug delivery, and controlled release with a particular emphasis on their most recent diagnostic and therapeutic applications.