The Scope, Methods and Applications of Biomedical Data Mining

Most literature assumptions have been drawn from public databases e.g. NHANES (National Health and Nutrition Examination Survey). Nonetheless, the sets of data are typically featured by high-dimensional timeliness, heterogeneity, characteristics and irregularity, hence amounting to valuation of these databases not being applied completely. Data Mining (DM) technologies have been the frontiers domains in biomedical studies, as it shows smart routine in assessing patients’ risks and aiding in the process of biomedical research and decision-making in developing disease-forecasting frameworks. In that case, DM has novel merits in biomedical Big Data (BD) studies, mostly in large-scale biomedical datasets. In this paper, a description of DM techniques alongside their fundamental practical applications will be provided. The objectives of this study are to help biomedical researchers to attain intuitive and clear appreciative of the applications of data-mining technologies on biomedical BD to enhance to creation of biomedical results, which are relevant in a biomedical setting.

Concentration and quantification of Tilapia tilapinevirus from water using a simple iron flocculation coupled with probe-based RT-qPCR

Tilapia tilapinevirus (also known as tilapia lake virus, TiLV) is an important virus responsible for die-off of farmed tilapia globally. Detection and quantification of the virus from environmental DNA/RNA (eDNA/eRNA) using pond water represents a potential, noninvasive routine approach for pathogen monitoring and early disease forecasting in aquaculture systems. Here, we report a simple iron flocculation method for viral concentration from water combined with a newly developed hydrolysis probe quantitative RT-qPCR method for detection and quantification of TiLV. The RT-qPCR method targeting a conserved region of TiLV genome segment 9 has a detection limit of 10 viral copies per uL of template. The method had a 100% analytical specificity and sensitivity for TiLV. The optimized iron flocculation method was able to recover 16.11 +/- 3.3% of virus from water samples spiked with viral cultures. During disease outbreak cases from an open-caged system and a closed hatchery system, both tilapia and water samples were collected for detection and quantification of TiLV. The results revealed that TiLV was detected from both clinically sick fish and asymptomatic fish. Most importantly, the virus was successfully detected from water samples collected from different locations in the affected farms e.g. river water samples from affected cages (8.50 x 102 to 2.79 x 104 copies/L) and fish-rearing water samples, sewage, and reservoir (4.29 x 102 to 3.53 x 103 copies/L) from affected and unaffected ponds of the hatchery. In summary, this study suggests that the eRNA detection system using iron flocculation coupled with probe based-RT-qPCR is feasible for concentration and quantification of TiLV from water. This approach might be useful for noninvasive monitoring of TiLV in tilapia aquaculture systems and facilitating appropriate decisions on biosecurity interventions needed.

Wheat Blast: A Disease Spreading by Intercontinental Jumps and Its Management Strategies

Wheat blast (WB) caused by Magnaporthe oryzae pathotype Triticum (MoT) is an important fungal disease in tropical and subtropical wheat production regions. The disease was initially identified in Brazil in 1985, and it subsequently spread to some major wheat-producing areas of the country as well as several South American countries such as Bolivia, Paraguay, and Argentina. In recent years, WB has been introduced to Bangladesh and Zambia via international wheat trade, threatening wheat production in South Asia and Southern Africa with the possible further spreading in these two continents. Resistance source is mostly limited to 2NS carriers, which are being eroded by newly emerged MoT isolates, demonstrating an urgent need for identification and utilization of non-2NS resistance sources. Fungicides are also being heavily relied on to manage WB that resulted in increasing fungal resistance, which should be addressed by utilization of new fungicides or rotating different fungicides. Additionally, quarantine measures, cultural practices, non-fungicidal chemical treatment, disease forecasting, biocontrol etc., are also effective components of integrated WB management, which could be used in combination with varietal resistance and fungicides to obtain reasonable management of this disease.

The predictive skill of convolutional neural networks models for disease forecasting

In this paper we investigate the utility of one-dimensional convolutional neural network (CNN) models in epidemiological forecasting. Deep learning models, in particular variants of recurrent neural networks (RNNs) have been studied for ILI (Influenza-Like Illness) forecasting, and have achieved a higher forecasting skill compared to conventional models such as ARIMA. In this study, we adapt two neural networks that employ one-dimensional temporal convolutional layers as a primary building block—temporal convolutional networks and simple neural attentive meta-learners—for epidemiological forecasting. We then test them with influenza data from the US collected over 2010-2019. We find that epidemiological forecasting with CNNs is feasible, and their forecasting skill is comparable to, and at times, superior to, plain RNNs. Thus CNNs and RNNs bring the power of nonlinear transformations to purely data-driven epidemiological models, a capability that heretofore has been limited to more elaborate mechanistic/compartmental disease models.

FOCUS: Forecasting COVID-19 in the United States

Infectious disease forecasting has been a useful tool for public health planning and messaging during the COVID-19 pandemic. In partnership with the CDC, the organizers of the COVID-19 Forecast Hub have created a mechanism for forecasters from academia, industry, and government organizations to submit weekly near-term predictions of COVID-19 targets in the United States. Here we describe our efforts to participate in the COVID-19 Forecast Hub through the Forecasting COVID-19 in the United States (FOCUS) project. The effort led to more than three months of weekly submissions and development of an automated pipeline to generate forecasts. The models used in FOCUS yielded forecasts that ranked relatively well in terms of precision and accuracy.