Connective Tissue Growth Factor: Regulation, Diseases, and Drug Discovery.

In drug discovery, selecting targeted molecules is crucial as the target could directly affect drug efficacy and the treatment outcomes. As a member of the CCN family, CTGF (also known as CCN2) is an essential regulator in the progression of various diseases, including fibrosis, cancer, neurological disorders, and eye diseases. Understanding the regulatory mechanisms of CTGF in different diseases may contribute to the discovery of novel drug candidates. Summarizing the CTGF-targeting and -inhibitory drugs is also beneficial for the analysis of the efficacy, applications, and limitations of these drugs in different disease models. Therefore, we reviewed the CTGF structure, the regulatory mechanisms in various diseases, and drug development in order to provide more references for future drug discovery.

Never-Ending Learning for Explainable Brain Computing.

Exploring the nature of human intelligence and behavior is a longstanding pursuit in cognitive neuroscience, driven by the accumulation of knowledge, information, and data across various studies. However, achieving a unified and transparent interpretation of findings presents formidable challenges. In response, an explainable brain computing framework is proposed that employs the never-ending learning paradigm, integrating evidence combination and fusion computing within a Knowledge-Information-Data (KID) architecture. The framework supports continuous brain cognition investigation, utilizing joint knowledge-driven forward inference and data-driven reverse inference, bolstered by the pre-trained language modeling techniques and the human-in-the-loop mechanisms. In particular, it incorporates internal evidence learning through multi-task functional neuroimaging analyses and external evidence learning via topic modeling of published neuroimaging studies, all of which involve human interactions at different stages. Based on two case studies, the intricate uncertainty surrounding brain localization in human reasoning is revealed. The present study also highlights the potential of systematization to advance explainable brain computing, offering a finer-grained understanding of brain activity patterns related to human intelligence.

Biochar improves the humification process during pig manure composting: Insights into roles of the bacterial community and metabolic functions.

Biochar could promote humification in composting, nevertheless, its mechanism has not been fully explored from the perspective of the overall bacterial community and its metabolism. This study investigated the effects of bamboo charcoal (BC) and wheat straw biochar (WSB) on the humic acid (HA) and fulvic acid (FA) contents during pig manure composting. The results showed that BC enhanced humification more than WSB, and significantly increased the HA content and HA/FA ratio. The bacterial community structure under BC differed from those under the other treatments, and BC increased the abundance of bacteria associated with the transformation of organic matter compared with the other treatments. Furthermore, biochar enhanced the metabolism of carbohydrates and amino acids in the thermophilic and cooling phases, especially BC. Through Mantel tests and network analysis, we found that HA was mainly related to carbon source metabolism and the bacterial community, and BC might change the interaction patterns among carbohydrates, amino acid metabolism, Bacillales, Clostridiales, and Lactobacillales with HA and FA to improve the humification process during composting. These results are important for understanding the mechanisms associated with the effects of biochar on humification during composting.

Interrelated feature selection from health surveys using domain knowledge graph.

Finding patterns among risk factors and chronic illness can suggest similar causes, provide guidance to improve healthy lifestyles, and give clues for possible treatments for outliers. Prior studies have typically isolated data challenges from single-disease datasets. However, the predictive power of multiple diseases is more helpful in establishing a healthy lifestyle than investigating one disease. Most studies typically focus on single-disease datasets; however, to ensure that health advice is generalized and contemporary, the features that predict the likelihood of many diseases can improve health advice effectiveness when considering the patient's point of view. We construct and present a novel knowledge-based qualitative method to remove redundant features from a dataset and redefine the outliers. The results of our trials upon five annual chronic disease health surveys demonstrate that our Knowledge Graph-based feature selection, when applied to many machine learning and deep learning multi-label classifiers, can improve classification performance. Our methodology is compatible with future directions, such as graph neural networks. It provides clinicians with an efficient process to select the most relevant health survey questions and responses regarding single or many human organ systems.

Identifying predictive biomarkers for repetitive transcranial magnetic stimulation response in depression patients with explainability.

Repetitive Transcranial Magnetic Stimulation (rTMS) is an evidence-based treatment for depression. However, the patterns of response to this treatment modality are inconsistent. Whilst many people see a significant reduction in the severity of their depression following rTMS treatment, some patients do not. To support and improve patient outcomes, recent work is exploring the possibility of using Machine Learning to predict rTMS treatment outcomes. Our proposed model is the first to combine functional magnetic resonance imaging (fMRI) connectivity with deep learning techniques to predict treatment outcomes before treatment starts. Furthermore, with the use of Explainable AI (XAI) techniques, we identify potential biomarkers that may discriminate between rTMS responders and non-responders. Our experiments utilize 200 runs of repeated bootstrap sampling on two rTMS datasets. We compare performances between our proposed feedforward deep neural network against existing methods, and compare the average accuracy, balanced accuracy and F1-score on a held-out test set. The results of these experiments show that our model outperforms existing methods with an average accuracy of 0.9423, balanced accuracy of 0.9423, and F1-score of 0.9461 in a sample of 61 patients. We found that functional connectivity measures between the Subgenual Anterior Cingulate Cortex and Centeral Opercular Cortex are a key determinant of rTMS treatment response. This knowledge provides psychiatrists with further information to explore the potential mechanisms of responses to rTMS treatment. Our developed prototype is ready to be deployed across large datasets in multiple centres and different countries.

The role of sclerostin in lipid and glucose metabolism disorders.

Lipid and glucose metabolism are critical for human activities, and their disorders can cause diabetes and obesity, two prevalent metabolic diseases. Studies suggest that the bone involved in lipid and glucose metabolism is emerging as an endocrine organ that regulates systemic metabolism through bone-derived molecules. Sclerostin, a protein mainly produced by osteocytes, has been therapeutically targeted by antibodies for treating osteoporosis owing to its ability to inhibit bone formation. Moreover, recent evidence indicates that sclerostin plays a role in lipid and glucose metabolism disorders. Although the effects of sclerostin on bone have been extensively examined and reviewed, its effects on systemic metabolism have not yet been well summarized. In this paper, we provide a systemic review of the effects of sclerostin on lipid and glucose metabolism based on in vitro and in vivo evidence, summarize the research progress on sclerostin, and prospect its potential manipulation for obesity and diabetes treatment.

Comparisons of oblique lumbar interbody fusion and transforaminal lumbar interbody fusion for degenerative spondylolisthesis: a prospective cohort study with a 2-year follow-up.

Objective: This study aimed to compare the clinical outcomes between oblique (OLIF) and transforaminal lumbar interbody fusion (TLIF) for patients with degenerative spondylolisthesis during a 2-year follow-up. Methods: Patients with symptomatic degenerative spondylolisthesis who underwent OLIF (OLIF group) or TLIF (TLIF group) were prospectively enrolled in the authors' hospital and followed up for 2 years. The primary outcomes were treatment effects [changes in visual analog score (VAS) and Oswestry disability index (ODI) from baseline] at 2 years after surgery; these were compared between two groups. Patient characteristics, radiographic parameters, fusion status, and complication rates were also compared. Results: In total, 45 patients were eligible for the OLIF group and 47 patients for the TLIF group. The rates of follow-up were 89% and 87% at 2 years, respectively. The comparisons of primary outcomes demonstrated no different changes in VAS-leg (OLIF, 3.4 vs. TLIF, 2.7), VAS-back (OLIF, 2.5 vs. TLIF, 2.1), and ODI (OLIF, 26.8 vs. TLIF, 30). The fusion rates were 86.1% in the TLIF group and 92.5% in the OLIF group at 2 years (P = 0.365). The OLIF group had less estimated blood loss (median, 200 ml) than the TLIF group (median, 300 ml) (P < 0.001). Greater restoration of disc height was obtained by OLIF (mean, 4.6 mm) than the TLIF group (mean, 1.3 mm) in the early postoperative period (P < 0.001). The subsidence rate was lower in the OLIF group than that in the TLIF group (17.5% vs. 38.9%, P = 0.037). The rates of total problematic complications were not different between the two groups (OLIF, 14.6% vs. TLIF, 26.2%, P = 0.192). Conclusion: OLIF did not show better clinical outcomes than TLIF for degenerative spondylolisthesis, except for lesser blood loss, greater disc height restoration, and lower subsidence rate.

Gynecological cancer prognosis using machine learning techniques: A systematic review of the last three decades (1990-2022).

OBJECTIVE: Many Computer Aided Prognostic (CAP) systems based on machine learning techniques have been proposed in the field of oncology. The objective of this systematic review was to assess and critically appraise the methodologies and approaches used in predicting the prognosis of gynecological cancers using CAPs. METHODS: Electronic databases were used to systematically search for studies utilizing machine learning methods in gynecological cancers. Study risk of bias (ROB) and applicability were assessed using the PROBAST tool. 139 studies met the inclusion criteria, of which 71 predicted outcomes for ovarian cancer patients, 41 predicted outcomes for cervical cancer patients, 28 predicted outcomes for uterine cancer patients, and 2 predicted outcomes for gynecological malignancies broadly. RESULTS: Random forest (22.30 %) and support vector machine (21.58 %) classifiers were used most commonly. Use of clinicopathological, genomic and radiomic data as predictors was observed in 48.20 %, 51.08 % and 17.27 % of studies, respectively, with some studies using multiple modalities. 21.58 % of studies were externally validated. Twenty-three individual studies compared ML and non-ML methods. Study quality was highly variable and methodologies, statistical reporting and outcome measures were inconsistent, preventing generalized commentary or meta-analysis of performance outcomes. CONCLUSION: There is significant variability in model development when prognosticating gynecological malignancies with respect to variable selection, machine learning (ML) methods and endpoint selection. This heterogeneity prevents meta-analysis and conclusions regarding the superiority of ML methods. Furthermore, PROBAST-mediated ROB and applicability analysis demonstrates concern for the translatability of existing models. This review identifies ways that this can be improved upon in future works to develop robust, clinically translatable models within this promising field.

Deep learning and machine learning in psychiatry: a survey of current progress in depression detection, diagnosis and treatment.

Informatics paradigms for brain and mental health research have seen significant advances in recent years. These developments can largely be attributed to the emergence of new technologies such as machine learning, deep learning, and artificial intelligence. Data-driven methods have the potential to support mental health care by providing more precise and personalised approaches to detection, diagnosis, and treatment of depression. In particular, precision psychiatry is an emerging field that utilises advanced computational techniques to achieve a more individualised approach to mental health care. This survey provides an overview of the ways in which artificial intelligence is currently being used to support precision psychiatry. Advanced algorithms are being used to support all phases of the treatment cycle. These systems have the potential to identify individuals suffering from mental health conditions, allowing them to receive the care they need and tailor treatments to individual patients who are mostly to benefit. Additionally, unsupervised learning techniques are breaking down existing discrete diagnostic categories and highlighting the vast disease heterogeneity observed within depression diagnoses. Artificial intelligence also provides the opportunity to shift towards evidence-based treatment prescription, moving away from existing methods based on group averages. However, our analysis suggests there are several limitations currently inhibiting the progress of data-driven paradigms in care. Significantly, none of the surveyed articles demonstrate empirically improved patient outcomes over existing methods. Furthermore, greater consideration needs to be given to uncertainty quantification, model validation, constructing interdisciplinary teams of researchers, improved access to diverse data and standardised definitions within the field. Empirical validation of computer algorithms via randomised control trials which demonstrate measurable improvement to patient outcomes are the next step in progressing models to clinical implementation.

Biochar enhanced organic matter transformation during pig manure composting: Roles of the cellulase activity and fungal community.

Fungal degradation of cellulose is a key step in the conversion of organic matter in composting. This study investigated the effects of adding 10% biochar (including, prepared from corn stalk and rape stalk corresponding to CSB and RSB) on organic matter transformation in composting and determined the role of cellulase and fungal communities in the conversion of organic matter. The results showed that biochar could enhance the conversion of organic matter, especially in RSB treatment. Biochar could increase cellulase activity, and RSB could enhance 33.78% and 30.70% the average activity of cellulase compared with the control and CSB treatments in the mesophilic to thermophilic phase, respectively. The results of high throughput sequencing demonstrated that Basidiomycota dominant in mesophilic phase, and Ascomycota dominant in other phases of composting. The redundancy analysis showed that Alternaria, Thermomycees, Aspergillus, Wallemia, and Melanocarpus might be the key fungi for the degradation of organic matter, and Fusarium, Penicillium, and Scopulariopsis may promote the conversion of organic matter. Network showed that the addition of RSB changed the interactions between fungal communities and organic matter transformation, and RSB treatment enriched members of Ascomycota related to organic matter transformation and cellulase activity. These results indicated that RSB improved organic matter conversion by enhancing the role of cellulase and fungal communities.

Genomic Profiling Reveals the Variant Landscape of Sporadic Parathyroid Adenomas in Chinese Population.

OBJECTIVE: To define somatic variants of parathyroid adenoma (PA) and to provide novel insights into the underlying molecular mechanism of sporadic PA. METHODS: Basic clinical characteristics and biochemical indices of 73 patients with PA were collected. Whole-exome sequencing was performed on matched tumor-constitutional DNA pairs to detect somatic alterations. Functional annotation was carried out by ingenuity pathway analysis afterward. The protein expression of the variant gene was confirmed by immunohistochemistry, and the relationship between genotype and phenotype was analyzed. RESULTS: Somatic variants were identified in 1549 genes, with an average of 69 variants per tumor (range, 13-2109; total, 9083). Several novel recurrent somatic variants were detected, such as KMT2D (15/73), MUC4 (14/73), POTEH (13/73), CD22 (12/73), HSPA2 (12/73), HCFC1 (11/73), MAGEA1 (11/73), and SLC4A3 (11/73), besides the previously reported PA-related genes, including MEN1 (11/73), CASR (6/73), MTOR (4/73), ASXL3 (3/73), FAT1 (3/73), ZFX (5/73), EZH1 (2/73), POT1 (2/73), and EZH2 (1/73). Among them, KMT2D might be the candidate driver gene of PA. Crucially, 5 patients carried somatic mutations in CDC73, showed an aggressive phenotype similar to that of parathyroid carcinoma (PC), and had a decreased expression of parafibromin. Pathway analysis of recurrent potential PA-associated driver variant genes revealed functional enrichments in the signaling pathway of Notch. CONCLUSION: Our study expanded the pathogenic variant spectrum of PA and indicated that KMT2D might be a novel candidate driver gene and be considered as a diagnostic biomarker for PA. Meanwhile, CDC73 mutations might be an early developmental event from PA to PC. The results provided insights into elucidating the pathogenesis of parathyroid tumorigenesis and a certain basis for clinical diagnosis and treatment.

Biochar reduces bioavailability of phosphorus during swine manure composting: Roles of phoD-harboring bacterial community.

The bioavailability of phosphorus is a vital index for evaluating the quality of compost products. This study examined the effects of adding wheat straw biochar (WSB) and bamboo charcoal (BC) on the transformation of various phosphorus fractions during composting, as well as analyzing the roles of the phoD-harboring bacterial community in the transformation of phosphorus fractions. Adding WSB and BC reduced the available phosphorus content in the compost products by 35.2 % and 38.5 %, respectively. Redundancy analysis showed that the alkaline phosphatase content and pH were the most important factors that affected the transformation of phosphorus fractions. The addition of biochar resulted in changes in the composition and structures of the phoD-harboring bacteria communities during composting. In addition, the key bacterial genera that secreted alkaline phosphatase and decomposed different forms of phosphorus under WSB and BC were different compared with those under control. Network and correlation analysis demonstrated that the activities of phoD-harboring bacteria could have been enhanced by biochar to accelerate the consumption of available phosphorus, and the activities of key phosphorus-solubilizing bacteria (Lysobacter, Methylobacterium, and Saccharothrix) might be inhibited when the pH increased, thereby increasing the insoluble phosphorus content.

Identifying informative tweets during a pandemic via a topic-aware neural language model.

Every epidemic affects the real lives of many people around the world and leads to terrible consequences. Recently, many tweets about the COVID-19 pandemic have been shared publicly on social media platforms. The analysis of these tweets is helpful for emergency response organizations to prioritize their tasks and make better decisions. However, most of these tweets are non-informative, which is a challenge for establishing an automated system to detect useful information in social media. Furthermore, existing methods ignore unlabeled data and topic background knowledge, which can provide additional semantic information. In this paper, we propose a novel Topic-Aware BERT (TABERT) model to solve the above challenges. TABERT first leverages a topic model to extract the latent topics of tweets. Secondly, a flexible framework is used to combine topic information with the output of BERT. Finally, we adopt adversarial training to achieve semi-supervised learning, and a large amount of unlabeled data can be used to improve inner representations of the model. Experimental results on the dataset of COVID-19 English tweets show that our model outperforms classic and state-of-the-art baselines.

Case report: Clinical characteristics and treatment of secondary osteoporosis induced by X-linked congenital adrenal dysplasia.

Objective: To summarize the clinical features and bone complications in a patient from a large family with X-linked congenital adrenocortical hypoplasia (AHC) and evaluate the efficacy of different treatment regimens on the prognosis of secondary osteoporosis caused by AHC at a 5-year follow-up. Methods: A large family with AHC was recruited, and the causative gene mutation was identified by Sanger sequencing in the proband. Clinical features as well as radiological examinations and laboratory indices of osteoporosis secondary to AHC were analyzed in this study. Meanwhile, the proband was treated with classical antiresorptive drugs (bisphosphonates) for 2 years and switched to a vitamin K2 analogue for another 3 years, during which the efficacy of the drugs was evaluated. Results: The proband was identified as carrying a homozygous insertion mutation (p. Thr193GlyfsX13) in the NR0B1 (nuclear receptor subfamily 0, group B, member 1) gene, resulting in a premature stop codon due to a frameshift mutation. During treatment and follow-up, the proband did not respond well to bisphosphonate and developed atypical femoral fractures. Vitamin K2 improved clinical symptoms. In terms of bone mineral density (BMD), there is no evidence of any effect of vitamin K2 on the neck of femur, though some minor effects on spinal BMD cannot be excluded. Conclusions: Secondary osteoporosis induced by AHC deserves clinical attention. Unlike in primary osteoporosis, the curative effect of bisphosphonates was unsatisfactory and was more likely to cause atypical femoral fractures in long-term treatment. It is suggested that bone anabolic drugs may be better alternatives.

Clinical and genetic characteristics of 29 Chinese patients with X-linked hypophosphatemia.

Objective: The aim of this study was to fully describe the clinical and genetic characteristics, including clinical manifestations, intact fibroblast growth factor 23 (iFGF23) levels, and presence of PHEX gene mutations, of 22 and 7 patients with familial and sporadic X-linked dominant hypophosphatemia (XLH), respectively. Methods: Demographic data, clinical features, biochemical indicators, and imaging data of 29 patients were collected. All 22 exons and exon-intron boundaries of the PHEX gene were amplified by polymerase chain reaction (PCR) and directly sequenced. The serum level of iFGF23 was measured in 15 of the patients. Results: Twenty-nine patients (male/female: 13:16, juvenile/adult: 15:14) with XLH were included. The main symptoms were bowed lower extremities (89.7%), abnormal gait (89.7%), and short stature/growth retardation (78.6%). Hypophosphatemia with a high alkaline phosphatase level was the main biochemical feature and the median value of serum iFGF23 was 55.7 pg/ml (reference range: 16.1-42.2 pg/ml). Eight novel mutations in the PHEX gene were identified by Sanger sequencing, including two missense mutations (p. Gln682Leu and p. Phe312Ser), two deletions (c.350_356del and c.755_761del), one insertion (c.1985_1986insTGAC), and three splice mutations (c.1700+5G>C, c.1966-1G>T, and c.350-14_350-1del). Additionally, the recurrence rate after the first orthopedic surgery was 77.8% (7/9), and five of them had their first surgery before puberty. Conclusion: Our study expanded the clinical phenotypes and gene mutation spectrum of XLH and provided a reference for the optimal timing of orthopedic surgeries.

Triaging Medical Referrals Based on Clinical Prioritisation Criteria Using Machine Learning Techniques.

Triaging of medical referrals can be completed using various machine learning techniques, but trained models with historical datasets may not be relevant as the clinical criteria for triaging are regularly updated and changed. This paper proposes the use of machine learning techniques coupled with the clinical prioritisation criteria (CPC) of Queensland (QLD), Australia, to deliver better triaging for referrals in accordance with the CPC's updates. The unique feature of the proposed model is its non-reliance on the past datasets for model training. Medical Natural Language Processing (NLP) was applied in the proposed approach to process the medical referrals, which are unstructured free text. The proposed multiclass classification approach achieved a Micro F1 score = 0.98. The proposed approach can help in the processing of two million referrals that the QLD health service receives annually; therefore, they can deliver better and more efficient health services.

Clinical Characteristics and Pathogenic Gene Identification in Chinese Patients With Paget's Disease of Bone.

Objective: To evaluate the clinical features of sporadic Paget's disease of bone (PDB) in China and further explore the underlying genetic abnormalities of the disease. Methods: Clinical characteristics, biochemical indices, bone turnover markers and radiographic examinations of the patients were collected. Genomic DNA was extracted from peripheral blood and whole-exome sequencing was carried out to identify the potential pathogenic genes. The pathogenicity of the variants was thereafter investigated by bioinformatics analysis. Results: A total of 50 patients (57.20 ± 15.52 years, male/female: 1.63: 1) with PDB were included and the mean onset age was 48.34 years (48.34 ± 17.24 years). 94.0% of the patients exhibited symptomatic patterns described as bone pain (86.0%), elevated skin temperature at the lesion site (26.0%), bone deformity (22.0%) and local swelling (18.0%). The most frequently involved lesion sites were pelvis (52.0%), femur (42.0%), tibia (28.0%), skull (28.0%) and spine (18.0%), respectively. Additionally, 40.0% of them accompanied with osteoarthritis, 14.0% with pathological fractures, and the misdiagnosis rate of PDB was as high as 36.0%. Serum level of alkaline phosphatase was significantly increased, with the mean value of 284.00 U/L (quartiles, 177.00-595.00 U/L). Two heterozygous missense mutations of SQSTM1 gene (c.1211T>C, M404T) and one novel heterozygous missense mutation in HNRNPA2B1 gene (c.989C>T, p. P330L) were identified in our study. Moreover, several potential disease-causing genes were detected and markedly enriched in the pathways of neurodegeneration (including WNT16, RYR3 and RYR1 genes) and amyotrophic lateral sclerosis (ALS, including NUP205, CAPN2, and NUP214 genes). Conclusion: In contrast to Western patients, Chinese patients have an earlier onset age, more severe symptoms, and lower frequency of SQSTM1 gene mutation (4.0%). Moreover, a novel heterozygous missense mutation in HNRNPA2B1 gene was identified in one male patient with isolated bone phenotype. As for other genetic factors, it was indicated that WNT16, RYR3, RYR1, NUP205, CAPN2 and NUP214 genes may be potential pathogenic genes, pathways of neurodegeneration and ALS may play a vital role in the pathogenesis of PDB.

Clinical characteristics and identification of a novel TGFB1 variant in three unrelated Chinese families with Camurati-Engelmann disease.

BACKGROUND: To investigate the clinical characteristics and molecular diagnosis of Camurati-Engelmann disease (CAEND) in Chinese individuals. METHODS: We recruited six patients aged 14 to 45 years in three unrelated families with CAEND, including five females and one male. Clinical manifestations, biochemical tests, and radiographic examinations were analyzed. The TGFB1 gene variants were further identified by Sanger sequencing. In addition, one female patient was followed up for 5 years. RESULTS: The onset age of the patients ranged from 1 to 6 years. All of them had family histories and consisted of an autosomal dominant inheritance pattern. Gait disturbance, fatigue, progressive bone pain, muscle atrophy, and weakness were the main complaints. Laboratory examinations revealed that the inflammatory markers were at high levels, in addition to the increased bone metabolism indicators. The thickened diaphysis of long bones and the narrowed medullary cavity was observed by radiography. Furthermore, bone scintigraphy detected abnormal symmetrical radioactive concentrations in the affected regions of bone. Sanger sequencing identified a missense heterozygous variant in exon 4 of the TGFB1 gene in families 1 and 2, resulting in Arg218Cys, which confirmed CAEND. Moreover, one novel variant c.669C > G in exon 4 of the TGFB1 gene harboring Cys223Trp was detected in family 3. Subsequent bioinformatics software predicted that the novel variant was pathogenic. Of interest, III:2 in family 3 experienced heart valve defects and tachycardia at birth, which had never been reported in CAEND patients before. Moreover, the response to drug treatment is also full of contradictions and is worthy of further study. CONCLUSION: Besides the typical CAEND manifestations, the new phenotypic characteristics of tachycardia and heart valve defects were first reported in one woman carrying the novel variant p.Cys223Trp in TGFB1 the gene. In addition, we demonstrated that increased bone metabolism indicators and inflammatory markers may possess auxiliary diagnosis for CAEND.

Machine and cognitive intelligence for human health: systematic review.

Brain informatics is a novel interdisciplinary area that focuses on scientifically studying the mechanisms of human brain information processing by integrating experimental cognitive neuroscience with advanced Web intelligence-centered information technologies. Web intelligence, which aims to understand the computational, cognitive, physical, and social foundations of the future Web, has attracted increasing attention to facilitate the study of brain informatics to promote human health. A large number of articles created in the recent few years are proof of the investment in Web intelligence-assisted human health. This study systematically reviews academic studies regarding article trends, top journals, subjects, countries/regions, and institutions, study design, artificial intelligence technologies, clinical tasks, and performance evaluation. Results indicate that literature is especially welcomed in subjects such as medical informatics and health care sciences and service. There are several promising topics, for example, random forests, support vector machines, and conventional neural networks for disease detection and diagnosis, semantic Web, ontology mining, and topic modeling for clinical or biomedical text mining, artificial neural networks and logistic regression for prediction, and convolutional neural networks and support vector machines for monitoring and classification. Additionally, future research should focus on algorithm innovations, additional information use, functionality improvement, model and system generalization, scalability, evaluation, and automation, data acquirement and quality improvement, and allowing interaction. The findings of this study help better understand what and how Web intelligence can be applied to promote healthcare procedures and clinical outcomes. This provides important insights into the effective use of Web intelligence to support informatics-enabled brain studies.

Streamflow prediction using an integrated methodology based on convolutional neural network and long short-term memory networks.

Streamflow (Qflow) prediction is one of the essential steps for the reliable and robust water resources planning and management. It is highly vital for hydropower operation, agricultural planning, and flood control. In this study, the convolution neural network (CNN) and Long-Short-term Memory network (LSTM) are combined to make a new integrated model called CNN-LSTM to predict the hourly Qflow (short-term) at Brisbane River and Teewah Creek, Australia. The CNN layers were used to extract the features of Qflow time-series, while the LSTM networks use these features from CNN for Qflow time series prediction. The proposed CNN-LSTM model is benchmarked against the standalone model CNN, LSTM, and Deep Neural Network models and several conventional artificial intelligence (AI) models. Qflow prediction is conducted for different time intervals with the length of 1-Week, 2-Weeks, 4-Weeks, and 9-Months, respectively. With the help of different performance metrics and graphical analysis visualization, the experimental results reveal that with small residual error between the actual and predicted Qflow, the CNN-LSTM model outperforms all the benchmarked conventional AI models as well as ensemble models for all the time intervals. With 84% of Qflow prediction error below the range of 0.05 m3 s-1, CNN-LSTM demonstrates a better performance compared to 80% and 66% for LSTM and DNN, respectively. In summary, the results reveal that the proposed CNN-LSTM model based on the novel framework yields more accurate predictions. Thus, CNN-LSTM has significant practical value in Qflow prediction.