Search for: All records

We propose a novel pipeline for the real-time detection of myocardial infarction from a single heartbeat of a 12-lead electrocardiograms. We do so by merging a real-time R-spike detection algorithm with a deep learning Long-Short Term Memory (LSTM) network-based classifier. A comparative assessment of the classification performance of the resulting system is performed and provided. The proposed algorithm achieves an inter-patient classification accuracy of 95.76% (with a 95% Confidence Interval (CI) of ±2.4%), a recall of 96.67% (±2.4% 95% CI), specificity of 93.64% (±5.7% 95% CI), and the average J-Score is 90.31% (±6.2% 95% CI). These state-of-the-art myocardial infarction detection metrics are extremely promising and could pave the wave for the early detection of myocardial infarctions. This high accuracy is achieved with a processing time of 40 milliseconds, which is most appropriate for online classification as the time between fast heartbeats is around 300 milliseconds. 

Using electroencephalography (EEG) data from epileptic patients 1 , we investigated and compared functional connectivity networks of three various types of epileptiform discharges (ED; single, complex & repetitive spikes) in 4 regions of the brain. Our results showed different connectivity patterns among three ED types within-and between-brain regions. The one-way ANOVA test indicated significant differences between the mean of the average connectivity matrices (ACMs) of the single spike, which characterize focal epilepsy, and the other two ED types (complex & repetitive) which characterize generalized epilepsy. The interictal EEG segments, through the connectivity patterns they yield, could be considered as one of the key indicators for the diagnosis of focal or generalized epilepsy. 

Non-stationarity is often observed in Geographic datasets. One way to explain non-stationarity is to think of it as a hidden "local knowledge" that varies across space. It is inherently difficult to model such data as models built for one region do not necessarily fit another area as the local knowledge could be different. A solution for this problem is to construct multiple local models at various locations, with each local model accounting for a sub-region within which the data remains relatively stationary. However, this approach is sensitive to the size of data, as the local models are only trained from a subset of observations from a particular region. In this paper, we present a novel approach that addresses this problem by aggregating spatially similar sub-regions into relatively large partitions. Our insight is that although local knowledge shifts over space, it is possible for multiple regions to share the same local knowledge. Data from these regions can be aggregated to train a more accurate model. Experiments show that this method can handle non-stationary and outperforms when the dataset is relatively small. 

This study introduces a new multimodal deep regression method to predict cognitive test score in a 5-year longitudinal study on Alzheimer’s disease (AD). The proposed model takes advantage of multimodal data that includes cerebrospinal fluid (CSF) levels of tau and beta-amyloid, structural measures from magnetic resonance imaging (MRI), functional and metabolic measures from positron emission tomography (PET), and cognitive scores from neuropsychological tests (Cog), all with the aim of achieving highly accurate predictions of future Mini-Mental State Examination (MMSE) test scores up to five years after baseline biomarker collection. A novel data augmentation technique is leveraged to increase the numbers of training samples without relying on synthetic data. With the proposed method, the best and most encompassing regressor is shown to achieve better than state-of-the-art correlations of 85.07%(SD=1.59) for 6 months in the future, 87.39% (SD =1.48) for 12 months, 84.78% (SD=2.66) for 18 months, 85.13% (SD=2.19) for 24 months, 81.15% (SD=5.48) for 30 months, 81.17% (SD=4.44) for 36 months, 79.25% (SD=5.85) for 42 months, 78.98% (SD=5.79) for 48 months, 78.93%(SD=5.76) for 54 months, and 74.96% (SD=7.54) for 60 months. 

Recent years saw explosive growth of Human Geography Data, in which spatial non-stationarity is often observed, i.e., relationships between features depend on the location. For these datasets, a single global model cannot accurately describe the relationships among features that vary across space. To address this problem, a viable solution- that has been adopted by many studies-is to create multiple local models instead of a global one, with each local model representing a subregion of the space. However, the challenge with this approach is that the local models are only fitted to nearby observations. For sparsely sampled regions, the data could be too few to generate any high-quality model. This is especially true for Human Geography datasets, as human activities tend to cluster at a few locations. In this paper, we present a modeling method that addresses this problem by letting local models operate within relatively large subregions, where overlapping is allowed. Results from all local models are then fused using an inverse distance weighted approach, to minimize the impact brought by overlapping. Experiments showed that this method handles non-stationary geographic data very Well, even When they are unevenly distributed.