Data-Efficient, Chemistry-Aware Machine Learning Predictions of Diels-Alder Reaction Outcomes.

The application of machine learning models to the prediction of reaction outcomes currently needs large and/or highly featurized data sets. We show that a chemistry-aware model, NERF, which mimics the bonding changes that occur during reactions, allows for highly accurate predictions of the outcomes of Diels-Alder reactions using a relatively small training set, with no pretraining and no additional features. We establish a diverse data set of 9537 intramolecular, hetero-, aromatic, and inverse electron demand Diels-Alder reactions. This data set is used to train a NERF model, and the performance is compared against state-of-the-art classification and generative machine learning models across low- and high-data regimes, with and without pretraining. The predictive accuracy (regio- and site selectivity in the major product) achieved by NERF exceeds 90% when as little as 40% of the data set is used for training. Another high-performing model, Chemformer, requires a larger training data set (>45%) and pretraining to reach 90% Top-1 accuracy. Accurate predictions of less-represented reaction subclasses, such as those involving heteroatomic or aromatic substrates, require higher percentages of training data. We also show how NERF can use small amounts of additional training data to quickly learn new systems and improve its overall understanding of reactivity. Synthetic chemists stand to benefit as this model can be rapidly expanded and tailored to areas of chemistry corresponding to the low-data regime.

Cooperative driver pathway discovery via fusion of multi-relational data of genes, miRNAs and pathways.

Discovering driver pathways is an essential step to uncover the molecular mechanism underlying cancer and to explore precise treatments for cancer patients. However, due to the difficulties of mapping genes to pathways and the limited knowledge about pathway interactions, most previous work focus on identifying individual pathways. In practice, two (or even more) pathways interplay and often cooperatively trigger cancer. In this study, we proposed a new approach called CDPathway to discover cooperative driver pathways. First, CDPathway introduces a driver impact quantification function to quantify the driver weight of each gene. CDPathway assumes that genes with larger weights contribute more to the occurrence of the target disease and identifies them as candidate driver genes. Next, it constructs a heterogeneous network composed of genes, miRNAs and pathways nodes based on the known intra(inter)-relations between them and assigns the quantified driver weights to gene-pathway and gene-miRNA relational edges. To transfer driver impacts of genes to pathway interaction pairs, CDPathway collaboratively factorizes the weighted adjacency matrices of the heterogeneous network to explore the latent relations between genes, miRNAs and pathways. After this, it reconstructs the pathway interaction network and identifies the pathway pairs with maximal interactive and driver weights as cooperative driver pathways. Experimental results on the breast, uterine corpus endometrial carcinoma and ovarian cancer data from The Cancer Genome Atlas show that CDPathway can effectively identify candidate driver genes [area under the receiver operating characteristic curve (AUROC) of $\geq $0.9] and reconstruct the pathway interaction network (AUROC of>0.9), and it uncovers much more known (potential) driver genes than other competitive methods. In addition, CDPathway identifies 150% more driver pathways and 60% more potential cooperative driver pathways than the competing methods. The code of CDPathway is available at http://mlda.swu.edu.cn/codes.php?name=CDPathway.

Introducing activation functions into segmented regression model to address lag effects of interventions.

The interrupted time series (ITS) design is widely used to examine the effects of large-scale public health interventions and has the highest level of evidence validity. However, there is a notable gap regarding methods that account for lag effects of interventions.To address this, we introduced activation functions (ReLU and Sigmoid) to into the classic segmented regression (CSR) of the ITS design during the lag period. This led to the proposal of proposed an optimized segmented regression (OSR), namely, OSR-ReLU and OSR-Sig. To compare the performance of the models, we simulated data under multiple scenarios, including positive or negative impacts of interventions, linear or nonlinear lag patterns, different lag lengths, and different fluctuation degrees of the outcome time series. Based on the simulated data, we examined the bias, mean relative error (MRE), mean square error (MSE), mean width of the 95% confidence interval (CI), and coverage rate of the 95% CI for the long-term impact estimates of interventions among different models.OSR-ReLU and OSR-Sig yielded approximately unbiased estimates of the long-term impacts across all scenarios, whereas CSR did not. In terms of accuracy, OSR-ReLU and OSR-Sig outperformed CSR, exhibiting lower values in MRE and MSE. With increasing lag length, the optimized models provided robust estimates of long-term impacts. Regarding precision, OSR-ReLU and OSR-Sig surpassed CSR, demonstrating narrower mean widths of 95% CI and higher coverage rates.Our optimized models are powerful tools, as they can model the lag effects of interventions and provide more accurate and precise estimates of the long-term impact of interventions. The introduction of an activation function provides new ideas for improving of the CSR model.

DMIL-IsoFun: predicting isoform function using deep multi-instance learning.

MOTIVATION: Alternative splicing creates the considerable proteomic diversity and complexity on relatively limited genome. Proteoforms translated from alternatively spliced isoforms of a gene actually execute the biological functions of this gene, which reflect the functional knowledge of genes at a finer granular level. Recently, some computational approaches have been proposed to differentiate isoform functions using sequence and expression data. However, their performance is far from being desirable, mainly due to the imbalance and lack of annotations at isoform-level, and the difficulty of modeling gene-isoform relations. RESULT: We propose a deep multi-instance learning-based framework (DMIL-IsoFun) to differentiate the functions of isoforms. DMIL-IsoFun firstly introduces a multi-instance learning convolution neural network trained with isoform sequences and gene-level annotations to extract the feature vectors and initialize the annotations of isoforms, and then uses a class-imbalance Graph Convolution Network to refine the annotations of individual isoforms based on the isoform co-expression network and extracted features. Extensive experimental results show that DMIL-IsoFun improves the Smin and Fmax of state-of-the-art solutions by at least 29.6% and 40.8%. The effectiveness of DMIL-IsoFun is further confirmed on a testbed of human multiple-isoform genes, and maize isoforms related with photosynthesis. AVAILABILITY AND IMPLEMENTATION: The code and data are available at http://www.sdu-idea.cn/codes.php?name=DMIL-Isofun. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Hybrid Target Selections by "Hand Gestures + Facial Expression" for a Rehabilitation Robot.

In this study we propose a "hand gesture + face expression" human machine interaction technique, and apply this technique to bedridden rehabilitation robot. "Hand gesture + Facial expression" interactive technology combines the input mode of gesture and facial expression perception. It involves seven basic facial expressions that can be used to determine a target selecting task, while hand gestures are used to control a cursor's location. A controlled experiment was designed and conducted to evaluate the effectiveness of the proposed hybrid technology. A series of target selecting tasks with different target widths and layouts were designed to examine the recognition accuracy of hybrid control gestures. An interactive experiment applied to a rehabilitation robot is designed to verify the feasibility of this interactive technology applied to rehabilitation robots. The experimental results show that the "hand + facial expression" interactive gesture has strong robustness, which can provide a novel guideline for designing applications in VR interfaces, and it can be applied to the rehabilitation robots.

Precision hyperthermia-induced miRNA-409-3p upregulation inhibits migration, invasion, and EMT of gastric cancer cells by targeting KLF17.

BACKGROUND: Hyperthermic intraperitoneal chemotherapy (HIPEC) is widely used for clinical treatment of advanced cancers. However, the regulatory mechanism underlying precise hyperthermia treatment in advanced gastric cancer (AGC) remains unclear. MiR-409-3p is reportedly downregulated in a variety of cancers, although its role in regulating treatment of AGC by precise hyperthermia remains unclear. The underlying mechanisms of miRNA-medicated regulation have been investigated using predicted and validated miRNA-gene targets, confirming the role of miRNA in HIPEC; METHODS: We used quantitative real time PCR (qRT-PCR) to detect miR-409-3p expression in gastric cancer (GC), as well as adjacent normal tissues, following exposure to varying temperatures. We detected miR-409-3p targets using dual-luciferase assay, then performed cell apoptosis, western blotting, invasion, and migration assays to detect GC functions; RESULTS: MiR-409-3p was upregulated and downregulated in precise hyperthermia and AGC, respectively. Moreover, miR-409-3p upregulated the Krüppel-like-factor 17 (KLF17), which subsequently inhibited migration, invasiveness, and epithelial-mesenchymal transition (EMT) but promoted apoptosis in GC cells; CONCLUSIONS: Precise hyperthermia upregulated miR-409-3p and KLF17 indirectly, thereby inhibiting invasion, migration, and EMT, and promoting apoptosis of gastric cancer cells.

Differentiating isoform functions with collaborative matrix factorization.

MOTIVATION: Isoforms are alternatively spliced mRNAs of genes. They can be translated into different functional proteoforms, and thus greatly increase the functional diversity of protein variants (or proteoforms). Differentiating the functions of isoforms (or proteoforms) helps understanding the underlying pathology of various complex diseases at a deeper granularity. Since existing functional genomic databases uniformly record the annotations at the gene-level, and rarely record the annotations at the isoform-level, differentiating isoform functions is more challenging than the traditional gene-level function prediction. RESULTS: Several approaches have been proposed to differentiate the functions of isoforms. They generally follow the multi-instance learning paradigm by viewing each gene as a bag and the spliced isoforms as its instances, and push functions of bags onto instances. These approaches implicitly assume the collected annotations of genes are complete and only integrate multiple RNA-seq datasets. As such, they have compromised performance. We propose a data integrative solution (called DisoFun) to Differentiate isoform Functions with collaborative matrix factorization. DisoFun assumes the functional annotations of genes are aggregated from those of key isoforms. It collaboratively factorizes the isoform data matrix and gene-term data matrix (storing Gene Ontology annotations of genes) into low-rank matrices to simultaneously explore the latent key isoforms, and achieve function prediction by aggregating predictions to their originating genes. In addition, it leverages the PPI network and Gene Ontology structure to further coordinate the matrix factorization. Extensive experimental results show that DisoFun improves the area under the receiver operating characteristic curve and area under the precision-recall curve of existing solutions by at least 7.7 and 28.9%, respectively. We further investigate DisoFun on four exemplar genes (LMNA, ADAM15, BCL2L1 and CFLAR) with known functions at the isoform-level, and observed that DisoFun can differentiate functions of their isoforms with 90.5% accuracy. AVAILABILITY AND IMPLEMENTATION: The code of DisoFun is available at mlda.swu.edu.cn/codes.php?name=DisoFun. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Predicting functions of maize proteins using graph convolutional network.

BACKGROUND: Maize (Zea mays ssp. mays L.) is the most widely grown and yield crop in the world, as well as an important model organism for fundamental research of the function of genes. The functions of Maize proteins are annotated using the Gene Ontology (GO), which has more than 40000 terms and organizes GO terms in a direct acyclic graph (DAG). It is a huge challenge to accurately annotate relevant GO terms to a Maize protein from such a large number of candidate GO terms. Some deep learning models have been proposed to predict the protein function, but the effectiveness of these approaches is unsatisfactory. One major reason is that they inadequately utilize the GO hierarchy. RESULTS: To use the knowledge encoded in the GO hierarchy, we propose a deep Graph Convolutional Network (GCN) based model (DeepGOA) to predict GO annotations of proteins. DeepGOA firstly quantifies the correlations (or edges) between GO terms and updates the edge weights of the DAG by leveraging GO annotations and hierarchy, then learns the semantic representation and latent inter-relations of GO terms in the way by applying GCN on the updated DAG. Meanwhile, Convolutional Neural Network (CNN) is used to learn the feature representation of amino acid sequences with respect to the semantic representations. After that, DeepGOA computes the dot product of the two representations, which enable to train the whole network end-to-end coherently. Extensive experiments show that DeepGOA can effectively integrate GO structural information and amino acid information, and then annotates proteins accurately. CONCLUSIONS: Experiments on Maize PH207 inbred line and Human protein sequence dataset show that DeepGOA outperforms the state-of-the-art deep learning based methods. The ablation study proves that GCN can employ the knowledge of GO and boost the performance. Codes and datasets are available at http://mlda.swu.edu.cn/codes.php?name=DeepGOA .

Down-regulated expression of OPCML predicts an unfavorable prognosis and promotes disease progression in human gastric cancer.

BACKGROUND: OPCML belongs to the IgLON family of Ig domain-containing GPI-anchored cell adhesion molecules and was recently found to be involved in carcinogenesis, while its role in gastric cancer remains unclear. METHODS: We assessed expression and biological behavior of OPCML in gastric cancer. RESULTS: OPCML expression was markedly reduced in tumor tissues and cancer cell lines. Decreased OPCML expression had a significant association with unfavorable tumor stage (p = 0.007) and grading (p < 0.001). Furthermore, the results revealed that OPCML was an independent prognostic factor for overall survival in gastric cancer (p = 0.002). In addition, ectopic expression of OPCML in cancer cells significantly inhibited cell viability (p < 0.01) and colony formation (p < 0.001), arrest cell cycle in G0/G1 phase and induced apoptosis, and suppressed tumor formation in nude mice. The alterations of phosphorylation status of AKT and its substrate GSK3ß, up-regulation of pro-apoptotic regulators including caspase-3, caspase-9 and PARP, and up-regulation of cell cycle regulator p27, were implicated in the biological activity of OPCML in cancer cells. CONCLUSION: Down-regulated OPCML expression might serve as an independent predictor for unfavorable prognosis of patients, and the biological behavior supports its role as a tumor suppressor in gastric cancer.

Hyperthermic Intraperitoneal Perfusion Chemotherapy and Cytoreductive Surgery for Controlling Malignant Ascites From Ovarian Cancer.

BACKGROUND: Malignant ascites, a complication often seen in patients with ovarian cancer (OC), is difficult to treat, but hyperthermic intraperitoneal chemotherapy (HIPEC) has a good efficacy. OBJECTIVE: The aim of this study was to assess the efficacy of cytoreductive surgery (CRS) combined with HIPEC for controlling malignant ascites from OC. MATERIALS AND METHODS: From December 2009 until December 2014, 53 patients with OC and malignant ascites were treated with CRS and HIPEC. Patients in good health condition were treated with CRS followed by HIPEC (CRS + HIPEC), and patients in poor health condition were treated initially with B-mode ultrasound-guided HIPEC followed by delayed CRS upon improvement of their health condition (HIPEC + delayed CRS). Resolution of ascites, complete CRS, overall survival, and disease-free survival were analyzed. RESULTS: All patients showed ascites regression. The total objective remission rate was 100%, even for patients in the poor condition group before CRS. Complete CRS was successful in 30 (88.23%) of 34 patients in the good condition group, and 17 (89.47%) of 19 patients in the poor condition group (P > 0.05). Median disease-free survival and median overall survival were 21 and 39 months in the good condition group, and 22 and 38 months in the poor condition group, respectively (P > 0.05). CONCLUSIONS: Hyperthermic intraperitoneal chemotherapy is effective at controlling ascites in patients with OC, even for patients in poor condition before CRS, or when complete CRS is not feasible. Furthermore, the regression of ascites appears not to be dependent on complete resection.

Cytoreductive surgery combined with hyperthermic intraperitoneal chemoperfusion for the treatment of gastric cancer: A single-centre retrospective study.

AIM: Cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemoperfusion (HIPEC) is the treatment regime most likely to achieve prolonged survival in patients with peritoneal carcinomatosis from gastroenteric cancer. To date, few publications have focused on the treatment of patients with gastric cancer alone. Several controversies remain unsolved, including the safety and effectiveness of the CRS-HIPEC combination regime, particularly in cases where HIPEC is used as adjuvant treatment after CRS. Therefore, in the current study, we aimed to evaluate the safety and effectiveness of CRS combined with HIPEC in patients with gastric cancer. METHOD: Data from 231 patients with a median age of 55.1 years treated with the CRS-HIPEC combination regime between January 2009 and December 2014 were retrospectively reviewed. All patients underwent the combination therapy (mean of 2.4 cycles per patient, range, 1 to 4 cycles). RESULTS: Median overall survival was 37.0 months, with 1-, 2- and 3-year survival rates recorded as 83.4%, 68.5%, and 38.7%, respectively. The serum levels of carcinoembryonic antigen (CEA) and carbohydrate antigen 199 (CA199) were significantly decreased after combination therapy in the completeness of cytoreduction (CCR)-0 and CCR-1 groups, while no significant changes observed in marker levels were observed in the CC ≥2 group. The post-operative morbidity and mortality rates were 6.9% and 0.9%, respectively. Multivariate analysis revealed low TNM tumour stage, ascites condition and CCR score as independent predictors for better survival. CONCLUSION: In view of the acceptable morbidity and mortality rates we propose that CRS combined with HIPEC presents an effective and safe treatment modality for patients with gastric cancer, especially in cases where optimal cytoreduction is achieved before the HIPEC procedure.

Long noncoding RNA aberrant expression profiles after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy of AGC ascertained by microarray analysis.

Long noncoding RNAs (lncRNAs) have been shown to be involved in the development and progression of advanced gastric cancer (AGC). However, the roles of lncRNAs in advanced gastric cancer during the process of cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) are not well understood. A high-throughput microarray analysis was performed to compare the expression profiles of lncRNAs and messenger RNAs (mRNAs) in AGC serum samples during the process of CRS + HIPEC. Several potentially AGC-associated lncRNAs were verified by real-time quantitative reverse transcription polymerase chain reaction (PCR) analysis. Using abundant and varied probes, we were able to assess 33,045 lncRNAs and 30,215 mRNAs in our microarray. We found that 566 lncRNAs were differentially expressed (2-fold change) in AGC serum samples, indicating the significantly up- or downregulated lncRNAs play important roles in AGC during the process of CRS + HIPEC. Quantitative PCR results further verified that eight lncRNAs were aberrantly expressed in AGC serum samples after CRS + HIEC compared with matched serum sample before CRS + HIPEC. Among them, BC031243 and RP11-356I2.2 were the most aberrantly expressed lncRNAs, as estimated by quantitative PCR in six pairs of AGC serum samples. Our study demonstrated the expression patterns of lncRNAs in AGC serums before and after CRS + HIPEC by microarray. These results revealed that lncRNAs were differentially expressed during the process of CRS + HIPEC, suggesting that they might play key roles in tumor development.

Thermo-chemotherapy Induced miR-218 upregulation inhibits the invasion of gastric cancer via targeting Gli2 and E-cadherin.

Thermo-chemotherapy has been proven to reduce the invasion capability of cancer cells. However, the molecular mechanism underlying this anti-invasion effect is still unclear. In this study, the role of thermo-chemotherapy in the inhibition of tumor invasion was studied. The results demonstrated that expression of miR-218 was downregulated in gastric cancer tissues, which had a positive correlation with tumor invasion and metastasis. In vitro thermo-chemotherapy increased miR-218 expression in SGC7901 cells and inhibited both proliferation and invasion of cancer cells. Gli2 was identified as a downstream target of miR-218, and its expression was negatively regulated by miR-218. The thermo-chemotherapy induced miR-218 upregulation was also accompanied by increasing of E-cadherin expression. In conclusion, the present study indicates that thermo-chemotherapy can effectively decrease the invasion capability of cancer cells and increase cell-cell adhesion. miR-218 and its downstream target Gli2, as well as E-cadherin, participate in the anti-invasion process.

Gradient-Based Local Causal Structure Learning.

Finding the causal structure from a set of variables given observational data is a crucial task in many scientific areas. Most algorithms focus on discovering the global causal graph but few efforts have been made toward the local causal structure (LCS), which is of wide practical significance and easier to obtain. LCS learning faces the challenges of neighborhood determination and edge orientation. Available LCS algorithms build on conditional independence (CI) tests, they suffer the poor accuracy due to noises, various data generation mechanisms, and small-size samples of real-world applications, where CI tests do not work. In addition, they can only find the Markov equivalence class, leaving some edges undirected. In this article, we propose a GradieNt-based LCS learning approach (GraN-LCS) to determine neighbors and orient edges simultaneously in a gradient-descent way, and, thus, to explore LCS more accurately. GraN-LCS formulates the causal graph search as minimizing an acyclicity regularized score function, which can be optimized by efficient gradient-based solvers. GraN-LCS constructs a multilayer perceptron (MLP) to simultaneously fit all other variables with respect to a target variable and defines an acyclicity-constrained local recovery loss to promote the exploration of local graphs and to find out direct causes and effects of the target variable. To improve the efficacy, it applies preliminary neighborhood selection (PNS) to sketch the raw causal structure and further incorporates an l1 -norm-based feature selection on the first layer of MLP to reduce the scale of candidate variables and to pursue sparse weight matrix. GraN-LCS finally outputs LCS based on the sparse weighted adjacency matrix learned from MLPs. We conduct experiments on both synthetic and real-world datasets and verify its efficacy by comparing against state-of-the-art baselines. A detailed ablation study investigates the impact of key components of GraN-LCS and the results prove their contribution.

An IMU-Based Ground Reaction Force Estimation Method and Its Application in Walking Balance Assessment.

Walking is one of the most common daily movements of the human body. Therefore, quantitative evaluation of human walking has been commonly used to assist doctors in grasping the disease degree and rehabilitation process of patients in the clinic. Compared with the kinematic characteristics, the ground reaction force (GRF) during walking can directly reflect the dynamic characteristics of human walking. It can further help doctors understand the degree of muscle recovery and joint coordination of patients. This paper proposes a GRF estimation method based on the elastic elements and Newton-Euler equation hybrid driving GRF estimation method. Compared with the existing research, the innovations are as follows. 1) The hardware system consists of only two inertial measurement units (IMUs) placed on shanks. The acquisition of the overall motion characteristics of human walking is realized through the simplified four-link walking model and the thigh prediction method. 2) The method was validated not only on 10 healthy subjects but also on 11 Parkinson's patients and 10 stroke patients with normalized mean absolute errors (NMAEs) of 5.95%±1.32%, 6.09%±2.00%, 5.87%±1.59%. 3) This paper proposes a dynamic balance assessment method based on the acquired motion data and the estimated GRF. It evaluates the overall balance ability and fall risk at four key time points for all subjects recruited. Because of the low-cost system, ease of use, low motion interference and environmental constraints, and high estimation accuracy, the proposed GRF estimation method and walking balance automatic assessment have broad clinical value.

Hyperthermia inhibits the progression of gastric cancer by downregulating PLEK2/PD-L1 and possibly participates in immunomodulation.

BACKGROUND: Hyperthermia is used as an adjunctive treatment for gastric cancer; however, the corresponding antitumor mechanism remains unclear. OBJECTIVE: To investigate the expression of PLEK2 in gastric cancer and the mechanism by which hyperthermia inhibits gastric cancer progression and participating in immunomodulation. METHODS: PLEK2 was screened by combining microarray analysis with gene knockdown and proliferation assays. Analysis based on the TCGA database, GEPIA website, and detection of clinical samples was employed to investigate the expression and correlation of PLEK2 and PD-L1. Knockdown of the expression PLEK2, subsequent experiments including western blotting, RT-qPCR, cell functional assays, and flow cytometry were used to assess the effects on cell migration, invasion, viability, and apoptosis. Intervention with hyperthermia to explore its effects. To evaluate the impact on immunity by detecting T cell proliferation and the release of IFNγ, activated T cells were co-cultured with the target cells. RESULTS: Hyperthermia significantly reduced the expression of PLEK2 and PD-L1, while both were increased in gastric cancer. Knockdown of PLEK2 inhibited PD-L1 expression and significantly inhibited the proliferation, invasion, migration, and viability of gastric cancer cells. A decrease in PLEK2 expression promotes cell apoptosis. Although it cannot affect the proliferation of activated T cells, it can partially reverse IFNγ suppression. CONCLUSION: PLEK2 plays a promoting role in gastric cancer, and hyperthermia downregulates PLEK2/PD-L1, which further inhibits cell proliferation, invasion, and migration, promotes cell apoptosis, and possibly participates in immune regulation.

Multivariate comparison of B-ultrasound guided and laparoscopic continuous circulatory hyperthermic intraperitoneal perfusion chemotherapy for malignant ascites.

OBJECTIVE: Clinical efficacy of B-ultrasound-guided and laparoscopy-assisted continuous hyperthermic intraperitoneal perfusion chemotherapy (CHIPC) for treatment of malignant ascites was investigated. METHODS: Sixty-two patients with malignant ascites induced by ovarian or gastrointestinal cancers were randomly treated with B-ultrasound-guided CHIPC (therapeutic group) or laparoscopy-assisted CHIPC (control group) performed at the same center. Hospitalization costs and surgical duration were evaluated. Follow-up was conducted for 21 months with B-ultrasound or computed tomography at least once per month for assessment of ascites amount and tumor progression. Clinical efficacy was assessed by modified World Health Organization criteria. Survival time, Karnofsky performance score (KPS) of quality of life (QOL), and complications were recorded for all patients. RESULTS: Overall condition, primary disease type, and ascites amounts were comparable between groups. Significantly shorter mean duration of perfusion catheter placement (35 vs. 85 min) and mean hospitalization cost (36,000 vs. 55,000 ¥/patient) were observed in the therapeutic group than the control group (P < 0.01). Significantly different KPS scores were not observed before or after CHIPC (23.13 vs. 22.64 %) in both groups (P > 0.05). No significant differences in objective remission rates of malignant ascites (93.75 vs. 93.34 %), median survival times (9 vs. 8 months), or stamp hole metastasis rates (18.75 vs. 18.15 %) were observed between groups (P > 0.05). CONCLUSIONS: B-ultrasound-guided and laparoscopy-assisted CHIPC have similar clinical efficacy for improving QOL and prolonging patient survival. B-ultrasound-guided CHIPC may, however, shorten operation times and reduce hospitalization costs, making the treatment available to a broader patient population, although port hole metastasis remains an issue.

Quaternion Factorization Machines: A Lightweight Solution to Intricate Feature Interaction Modeling.

Due to the sparsity of available features in web-scale predictive analytics, combinatorial features become a crucial means for deriving accurate predictions. As a well-established approach, a factorization machine (FM) is capable of automatically learning high-order interactions among features to make predictions without the need for manual feature engineering. With the prominent development of deep neural networks (DNNs), there is a recent and ongoing trend of enhancing the expressiveness of FM-based models with DNNs. However, though better results are obtained with DNN-based FM variants, such performance gain is paid off by an enormous amount (usually millions) of excessive model parameters on top of the plain FM. Consequently, the heavy parameterization impedes the real-life practicality of those deep models, especially efficient deployment on resource-constrained Internet of Things (IoT) and edge devices. In this article, we move beyond the traditional real space where most deep FM-based models are defined and seek solutions from quaternion representations within the hypercomplex space. Specifically, we propose the quaternion factorization machine (QFM) and quaternion neural factorization machine (QNFM), which are two novel lightweight and memory-efficient quaternion-valued models for sparse predictive analytics. By introducing a brand new take on FM-based models with the notion of quaternion algebra, our models not only enable expressive inter-component feature interactions but also significantly reduce the parameter size due to lower degrees of freedom in the hypercomplex Hamilton product compared with real-valued matrix multiplication. Extensive experimental results on three large-scale datasets demonstrate that QFM achieves 4.36% performance improvement over the plain FM without introducing any extra parameters, while QNFM outperforms all baselines with up to two magnitudes' parameter size reduction in comparison to state-of-the-art peer methods.

Impact of COVID-19 on health services utilization in mainland China and its different regions based on S-ARIMA predictions.

Global health services are disrupted by the COVID-19 pandemic. We evaluated extent and duration of impacts of the pandemic on health services utilization in different economically developed regions of mainland China. Based on monthly health services utilization data in China, we used Seasonal Autoregressive Integrated Moving Average (S-ARIMA) models to predict outpatient and emergency department visits to hospitals (OEH visits) per capita without pandemic. The impacts were evaluated by three dimensions:1) absolute instant impacts were evaluated by difference between predicted and actual OEH visits per capita in February 2020 and relative instant impacts were the ratio of absolute impacts to baseline OEH visits per capita; 2) absolute and relative accumulative impacts from February 2020 to March 2021; 3) duration of impacts was estimated by time that actual OEH visits per capita returned to its predicted value. From February 2020 to March 2021, the COVID-19 pandemic reduced OEH visits by 0.4676 per capita, equivalent to 659,453,647 visits, corresponding to a decrease of 15.52% relative to the pre-pandemic average annual level in mainland China. The instant impacts in central, northeast, east and west China were 0.1279, 0.1265, 0.1215, and 0.0986 visits per capita, respectively; and corresponding relative impacts were 77.63%, 66.16%, 44.39%, and 50.57%, respectively. The accumulative impacts in northeast, east, west and central China were up to 0.5898, 0.4459, 0.3523, and 0.3324 visits per capita, respectively; and corresponding relative impacts were 23.72%, 12.53%, 13.91%, and 16.48%, respectively. The OEH visits per capita has returned back to predicted values within the first 2, 6, 9, 9 months for east, central, west and northeast China, respectively. Less economically developed areas were affected for a longer time. Safe and equitable access to health services, needs paying great attention especially for undeveloped areas.

Optimized segmented regression models for the transition period of intervention effects.

BACKGROUND: The interrupted time series (ITS) design is a widely used approach to examine the effects of interventions. However, the classic segmented regression (CSR) method, the most popular statistical technique for analyzing ITS data, may not be adequate when there is a transitional period between the pre- and post-intervention phases. METHODS: To address this issue and better capture the distribution patterns of intervention effects during the transition period, we propose using different cumulative distribution functions in the CSR model and developing corresponding optimized segmented regression (OSR) models. This study illustrates the application of OSR models to estimate the long-term impact of a national free delivery service policy intervention in Ethiopia. RESULTS: Regardless of the choice of transition length ([Formula: see text]) and distribution patterns of intervention effects, the OSR models outperformed the CSR model in terms of mean square error (MSE), indicating the existence of a transition period and the validity of our model's assumptions. However, the estimates of long-term impacts using OSR models are sensitive to the selection of L, highlighting the importance of reasonable parameter specification. We propose a data-driven approach to select the transition period length to address this issue. CONCLUSIONS: Overall, our OSR models provide a powerful tool for modeling intervention effects during the transition period, with a superior model fit and more accurate estimates of long-term impacts. Our study highlights the importance of appropriate statistical methods for analyzing ITS data and provides a useful framework for future research.