White learning methodology: A case study of cancer-related disease factors analysis in real-time PACS environment.

BACKGROUND AND OBJECTIVE: Bayesian network is a probabilistic model of which the prediction accuracy may not be one of the highest in the machine learning family. Deep learning (DL) on the other hand possess of higher predictive power than many other models. How reliable the result is, how it is deduced, how interpretable the prediction by DL mean to users, remain obscure. DL functions like a black box. As a result, many medical practitioners are reductant to use deep learning as the only tool for critical machine learning application, such as aiding tool for cancer diagnosis. METHODS: In this paper, a framework of white learning is being proposed which takes advantages of both black box learning and white box learning. Usually, black box learning will give a high standard of accuracy and white box learning will provide an explainable direct acyclic graph. According to our design, there are 3 stages of White Learning, loosely coupled WL, semi coupled WL and tightly coupled WL based on degree of fusion of the white box learning and black box learning. In our design, a case of loosely coupled WL is tested on breast cancer dataset. This approach uses deep learning and an incremental version of Naïve Bayes network. White learning is largely defied as a systemic fusion of machine learning models which result in an explainable Bayes network which could find out the hidden relations between features and class and deep learning which would give a higher accuracy of prediction than other algorithms. We designed a series of experiments for this loosely coupled WL model. RESULTS: The simulation results show that using WL compared to standard black-box deep learning, the levels of accuracy and kappa statistics could be enhanced up to 50%. The performance of WL seems more stable too in extreme conditions such as noise and high dimensional data. The relations by Bayesian network of WL are more concise and stronger in affinity too. CONCLUSION: The experiments results deliver positive signals that WL is possible to output both high classification accuracy and explainable relations graph between features and class.

Altered Brain Regional Homogeneity Following Contralateral Acupuncture at Quchi (LI 11) and Zusanli (ST 36) in Ischemic Stroke Patients with Left Hemiplegia: An fMRI Study.

OBJECTIVE: To study the effect of contralateral acupuncture (CAT) at acupoints of Quchi (LI 11) and Zusanli (ST 36) on the unaffected limbs of ischemic stroke patients with left hemiplegia based on regional homogeneity (ReHo) indices. METHODS: Ten ischemic stroke patients with left hemiplegia received CAT on right side at LI 11 and ST 36. Functional magnetic resonance imaging (fMRI) was performed before and after acupuncture. A ReHo analytical method was used to compare brain responses of patients before and after CAT operated by REST software. RESULTS: The stimulation at both LI 11 and ST 36 on the unaffected limbs produced significantly different neural activities. CAT elicited increased ReHo values at the right precentral gyrus and superior frontal gyrus, decreased ReHo value at right superior parietal lobule, left fusiform gyrus and left supplementary motor area. CONCLUSIONS: Acupuncture at one side could stimulate bilateral regions. CAT could evoke the gyrus which was possibly related to motor recovery from stroke. A promising indicator of neurobiological deficiencies could be represented by ReHo values in post-stroke patients.

A Survey of Data Mining and Deep Learning in Bioinformatics.

The fields of medicine science and health informatics have made great progress recently and have led to in-depth analytics that is demanded by generation, collection and accumulation of massive data. Meanwhile, we are entering a new period where novel technologies are starting to analyze and explore knowledge from tremendous amount of data, bringing limitless potential for information growth. One fact that cannot be ignored is that the techniques of machine learning and deep learning applications play a more significant role in the success of bioinformatics exploration from biological data point of view, and a linkage is emphasized and established to bridge these two data analytics techniques and bioinformatics in both industry and academia. This survey concentrates on the review of recent researches using data mining and deep learning approaches for analyzing the specific domain knowledge of bioinformatics. The authors give a brief but pithy summarization of numerous data mining algorithms used for preprocessing, classification and clustering as well as various optimized neural network architectures in deep learning methods, and their advantages and disadvantages in the practical applications are also discussed and compared in terms of their industrial usage. It is believed that in this review paper, valuable insights are provided for those who are dedicated to start using data analytics methods in bioinformatics.

Adaptive Swarm Balancing Algorithms for rare-event prediction in imbalanced healthcare data.

Clinical data analysis and forecasting have made substantial contributions to disease control, prevention and detection. However, such data usually suffer from highly imbalanced samples in class distributions. In this paper, we aim to formulate effective methods to rebalance binary imbalanced dataset, where the positive samples take up only the minority. We investigate two different meta-heuristic algorithms, particle swarm optimization and bat algorithm, and apply them to empower the effects of synthetic minority over-sampling technique (SMOTE) for pre-processing the datasets. One approach is to process the full dataset as a whole. The other is to split up the dataset and adaptively process it one segment at a time. The experimental results reported in this paper reveal that the performance improvements obtained by the former methods are not scalable to larger data scales. The latter methods, which we call Adaptive Swarm Balancing Algorithms, lead to significant efficiency and effectiveness improvements on large datasets while the first method is invalid. We also find it more consistent with the practice of the typical large imbalanced medical datasets. We further use the meta-heuristic algorithms to optimize two key parameters of SMOTE. The proposed methods lead to more credible performances of the classifier, and shortening the run time compared to brute-force method.

WITHDRAWN: Adaptive Swarm Balancing Algorithms for rare-event prediction in imbalanced healthcare data.

This article has been withdrawn at the request of the author(s) and/or editor. The Publisher apologizes for any inconvenience this may cause. The full Elsevier Policy on Article Withdrawal can be found at http://www.elsevier.com/locate/withdrawalpolicy.

A rat model of chronic gastric sensorimotor dysfunction resulting from transient neonatal gastric irritation.

BACKGROUND & AIMS: Although several pathophysiologic abnormalities have been noted in functional dyspepsia (FD), their pathogenesis is poorly understood. We hypothesized that chronic gastric hypersensitivity and gastric motor dysfunction seen in FD patients can be modeled in rats by transient gastric irritation during the neonatal period, a time of known neuronal vulnerability to long-term plasticity. METHODS: Ten-day-old male rats received 0.2 mL 0.1% iodoacetamide (IA) in 2% sucrose daily by oral gavages for 6 days; controls received 2% sucrose. Rats in both groups were then followed to adulthood (8-10 weeks) at which point behavioral, visceromotor, and great splanchnic nerve responses to graded gastric balloon distention (GD; 20-80 mm Hg) and gastric motor function were tested. RESULTS: IA-treated rats exhibited hypersensitivity to GD in a dose-dependent manner, as compared with the control group. The threshold of afferent nerve activation was lower and nerve responses to GD were significantly increased in IA-treated rats. Although IA-treated rats ingested food at a lower rate, gastric emptying was not significantly different between IA and control groups. However, gastric accommodation was significantly reduced in the IA group. No significant gastric pathology was seen in hypersensitive adult rats compared with controls. CONCLUSIONS: These studies demonstrate that gastric irritation in the neonatal period can result in chronic gastric hypersensitivity and gastric motor dysfunction in adults even in the absence of significant detectable gastric pathology. Our results offer insight into the pathogenesis of chronic functional dyspepsia and provide a potential model for further study to this important clinical problem.

Transforming growth factor-beta3 increases gap-junctional communication among folliculostellate cells to release basic fibroblast growth factor.

Folliculostellate (FS) cells are known to communicate with each other and with endocrine cells via gap junctions in the anterior pituitary. We investigated whether TGFbeta3 and estradiol, known to regulate FS cell production and secretion of basic fibroblast growth factor (bFGF), increases gap junctional communication to alter bFGF secretion from FS cells. FS cells in monolayer cultures were treated with TGFbeta3 or vehicle alone for 24 h and then microinjected with Lucifer Yellow and high-molecular-weight Texas Red dextran. Ten minutes later the transfer of dye among adjacent cells was recorded with a digital microscope. TGFbeta3 increased the transfer of dye. The TGFbeta3-neutralizing antibody and the gap junction inhibitor octanol reduced the effect of TGFbeta3 on the transfer of dye. The TGFbeta3-induced transfer of dye was unaltered by simultaneous treatment with estradiol. The steroid alone also had no effect. TGFbeta3 increased total and phosphorylated levels of connexin 43. Estradiol treatment did not produce any significant changes on basal or TGFbeta3-induced increases in connexin 43 levels. The gap-junction inhibitor octanol reduced TGFbeta3-increased levels of bFGF in FS cells. Taken together, these results suggest that TGFbeta3 may act on FS cells to increase gap-junctional communication to maximize its effect on bFGF secretion.

Oral uridine pro-drug PN401 decreases neurodegeneration, behavioral impairment, weight loss and mortality in the 3-nitropropionic acid mitochondrial toxin model of Huntington's disease.

Huntington's disease (HD) is associated with decreased activity of mitochondrial succinate dehydrogenase (complex II). De novo biosynthesis of uridine nucleotides is directly coupled to the respiratory chain. Cells with impaired mitochondrial function become uridine auxotrophs and can be maintained with high micromolar concentration of uridine and pyruvate. The therapeutic role of pyrimidines and possible changes in uridine content has not been assessed in neurological diseases involving mitochondrial dysfunction in vivo. Oral administration of PN401 delivers much higher levels of uridine to the circulation than oral administration of uridine itself. Administration of complex II inhibitor 3-nitropropionic acid (3NP) induced neuronal damage in the striatum, substantia nigra and/or thalamus in 80% of the mice and led to 38% mortality. Treatment with PN401 almost completely prevented the neuronal damage due to 3NP and completely prevented mortality. In two subsequent experiments, 3NP-induced weight loss, mortality and behavioral impairment in rotarod performance and spontaneous motor activity were attenuated by treatment with oral PN401. 3NP did not reduce forebrain total uridine nucleotides (TUN), though higher doses of PN401 associated with optimal neuroprotection did elevate TUN to supranormal levels. Thus, oral PN401 treatment has neuroprotective effects in a HD model of mitochondrial dysfunction and the mechanism is more complex than correction of a pyrimidine deficit.