In many lower-and-middle income countries including South Africa, data access in health facilities is restricted due to patient privacy and confidentiality policies. Further, since clinical data is unique to individual institutions and laboratories, there are insufficient data annotation standards and conventions. As a result of the scarcity of textual data, natural language processing (NLP) techniques have fared poorly in the health sector. A cause of death (COD) is often determined by a verbal autopsy (VA) report in places without reliable death registration systems. A non-clinician field worker does a VA report using a set of standardized questions as a guide to uncover symptoms of a COD. This analysis focuses on the textual part of the VA report as a case study to address the challenge of adapting NLP techniques in the health domain. We present a system that relies on two transfer learning paradigms of monolingual learning and multi-source domain adaptation to improve VA narratives for the target task of the COD classification. We use the Bidirectional Encoder Representations from Transformers (BERT) and Embeddings from Language Models (ELMo) models pre-trained on the general English and health domains to extract features from the VA narratives. Our findings suggest that this transfer learning system improves the COD classification tasks and that the narrative text contains valuable information for figuring out a COD. Our results further show that combining binary VA features and narrative text features learned via this framework boosts the classification task of COD.

Ethereum is one of the most valuable blockchain networks in terms of the total monetary value locked in it, and arguably been the most active network where new blockchain innovations in research and applications are demonstrated. But, this also leads to Ethereum network being susceptible to a wide variety of threats and attacks in an attempt to gain unreasonable advantage or to undermine the value of the users. Even with the state-of-art classical ML algorithms, detecting such attacks is still hard. This motivated us to build a hybrid system of quantum-classical algorithms that improves phishing detection in financial transaction networks. This paper presents a classical ensemble pipeline of classical and quantum algorithms and a detailed study benchmarking existing Quantum Machine Learning algorithms such as Quantum Support Vector Machine and Variational Quantum Classifier. With the current generation of quantum hardware available, smaller datasets are more suited to the QML models and most research restricts to hundreds of samples. However, we experimented on different data sizes and report results with a test data of 12K transaction nodes, which is to the best of the authors knowledge the largest QML experiment run so far on any real quantum hardware. The classical ensembles of quantum-classical models improved the macro F-score and phishing F-score. One key observation is QSVM constantly gives lower false positives, thereby higher precision compared with any other classical or quantum network, which is always preferred for any anomaly detection problem. This is true for QSVMs when used individually or via bagging of same models or in combination with other classical/quantum models making it the most advantageous quantum algorithm so far. The proposed ensemble framework is generic and can be applied for any classification task

In the literature of mitigating unfairness in machine learning, many fairness measures are designed to evaluate predictions of learning models and also utilised to guide the training of fair models. It has been theoretically and empirically shown that there exist conflicts and inconsistencies among accuracy and multiple fairness measures. Optimising one or several fairness measures may sacrifice or deteriorate other measures. Two key questions should be considered, how to simultaneously optimise accuracy and multiple fairness measures, and how to optimise all the considered fairness measures more effectively. In this paper, we view the mitigating unfairness problem as a multi-objective learning problem considering the conflicts among fairness measures. A multi-objective evolutionary learning framework is used to simultaneously optimise several metrics (including accuracy and multiple fairness measures) of machine learning models. Then, ensembles are constructed based on the learning models in order to automatically balance different metrics. Empirical results on eight well-known datasets demonstrate that compared with the state-of-the-art approaches for mitigating unfairness, our proposed algorithm can provide decision-makers with better tradeoffs among accuracy and multiple fairness metrics. Furthermore, the high-quality models generated by the framework can be used to construct an ensemble to automatically achieve a better tradeoff among all the considered fairness metrics than other ensemble methods. Our code is publicly available at https://github.com/qingquan63/FairEMOL

The goal of emotional brain state classification on functional MRI (fMRI) data is to recognize brain activity patterns related to specific emotion tasks performed by subjects during an experiment. Distinguishing emotional brain states from other brain states using fMRI data has proven to be challenging due to two factors: a difficulty to generate fast yet accurate predictions in short time frames, and a difficulty to extract emotion features which generalize to unseen subjects. To address these challenges, we conducted an experiment in which 22 subjects viewed pictures designed to stimulate either negative, neutral or rest emotional responses while their brain activity was measured using fMRI. We then developed two distinct Convolution-based approaches to decode emotional brain states using only spatial information from single, minimally pre-processed (slice timing and realignment) fMRI volumes. In our first approach, we trained a 1D Convolutional Network (84.9% accuracy; chance level 33%) to classify 3 emotion conditions using One-way Analysis of Variance (ANOVA) voxel selection combined with hyperalignment. In our second approach, we trained a 3D ResNet-50 model (78.0% accuracy; chance level 50%) to classify 2 emotion conditions from single 3D fMRI volumes directly. Our Convolutional and Residual classifiers successfully learned group-level emotion features and could decode emotion conditions from fMRI volumes in milliseconds. These approaches could potentially be used in brain computer interfaces and real-time fMRI neurofeedback research.

Interplanetary coronal mass ejections (ICMEs) are one of the main drivers for space weather disturbances. In the past, different approaches have been used to automatically detect events in existing time series resulting from solar wind in situ observations. However, accurate and fast detection still remains a challenge when facing the large amount of data from different instruments. For the automatic detection of ICMEs we propose a pipeline using a method that has recently proven successful in medical image segmentation. Comparing it to an existing method, we find that while achieving similar results, our model outperforms the baseline regarding training time by a factor of approximately 20, thus making it more applicable for other datasets. The method has been tested on in situ data from the Wind spacecraft between 1997 and 2015 with a True Skill Statistic (TSS) of 0.64. Out of the 640 ICMEs, 466 were detected correctly by our algorithm, producing a total of 254 False Positives. Additionally, it produced reasonable results on datasets with fewer features and smaller training sets from Wind, STEREO-A and STEREO-B with True Skill Statistics of 0.56, 0.57 and 0.53, respectively. Our pipeline manages to find the start of an ICME with a mean absolute error (MAE) of around 2 hours and 56 minutes, and the end time with a MAE of 3 hours and 20 minutes. The relatively fast training allows straightforward tuning of hyperparameters and could therefore easily be used to detect other structures and phenomena in solar wind data, such as corotating interaction regions.

We present fast classification techniques for sparse generalized linear and additive models. These techniques can handle thousands of features and thousands of observations in minutes, even in the presence of many highly correlated features. For fast sparse logistic regression, our computational speed-up over other best-subset search techniques owes to linear and quadratic surrogate cuts for the logistic loss that allow us to efficiently screen features for elimination, as well as use of a priority queue that favors a more uniform exploration of features. As an alternative to the logistic loss, we propose the exponential loss, which permits an analytical solution to the line search at each iteration. Our algorithms are generally 2 to 5 times faster than previous approaches. They produce interpretable models that have accuracy comparable to black box models on challenging datasets.

Brain Computer Interface (BCI) has great potential for solving many brain signal analysis limitations, mental disorder resolutions, and restoring missing limb functionality via neural-controlled implants. However, there is no single available, and safe implant for daily life usage exists yet. Most of the proposed implants have several implementation issues, such as infection hazards and heat dissipation, which limits their usability and makes it more challenging to pass regulations and quality control production. The wireless implant does not require a chronic wound in the skull. However, the current complex clustering neuron identification algorithms inside the implant chip consume a lot of power and bandwidth, causing higher heat dissipation issues and draining the implant's battery. The spike sorting is the core unit of an invasive BCI chip, which plays a significant role in power consumption, accuracy, and area. Therefore, in this study, we propose a low-power adaptive simplified VLSI architecture, "Zydeco-Style," for BCI spike sorting that is computationally less complex with higher accuracy that performs up to 93.5% in the worst-case scenario. The architecture uses a low-power Bluetooth Wireless communication module with external IoT medical ICU devices. The proposed architecture was implemented and simulated in Verilog. In addition, we are proposing an implant conceptual design.

Graph neural networks have emerged as a promising approach for the analysis of non-Euclidean data such as meshes. In medical imaging, mesh-like data plays an important role for modelling anatomical structures, and shape classification can be used in computer aided diagnosis and disease detection. However, with a plethora of options, the best architectural choices for medical shape analysis using GNNs remain unclear. We conduct a comparative analysis to provide practitioners with an overview of the current state-of-the-art in geometric deep learning for shape classification in neuroimaging. Using biological sex classification as a proof-of-concept task, we find that using FPFH as node features substantially improves GNN performance and generalisation to out-of-distribution data; we compare the performance of three alternative convolutional layers; and we reinforce the importance of data augmentation for graph based learning. We then confirm these results hold for a clinically relevant task, using the classification of Alzheimer's disease.

This page contains ML Fundamentals glossary terms. The number of correct classification predictions divided by the total number of predictions. Binary classification provides specific names for the different categories of correct predictions and incorrect predictions. Compare and contrast accuracy with precision and recall. Although a valuable metric for some situations, accuracy is highly misleading for others. Notably, accuracy is usually a poor metric for evaluating classification models that process class-imbalanced datasets. For example, suppose snow falls only 25 days per century in a certain subtropical city. Since days without snow (the negative class) vastly outnumber days with snow (the positive class), the snow dataset for this city is class-imbalanced. Imagine a binary classification model that is supposed to predict either snow or no snow each day but simply predicts "no snow" every day. This model is highly accurate but has no predictive power. Although 99.93% accuracy seems like very a impressive percentage, the model actually has no predictive power. Precision and recall are usually more useful metrics than accuracy for evaluating models trained on class-imbalanced datasets. A function that enables neural networks to learn nonlinear (complex) relationships between features and the label. The plots of activation functions are never single straight lines. In a neural network, activation functions manipulate the weighted sum of all the inputs to a neuron. To calculate a weighted sum, the neuron adds up the products of the relevant values and weights. A non-human program or model that can solve sophisticated tasks. For example, a program or model that translates text or a program or model that identifies diseases from radiologic images both exhibit artificial intelligence. Formally, machine learning is a sub-field of artificial intelligence. However, in recent years, some organizations have begun using the terms artificial intelligence and machine learning interchangeably. A number between 0.0 and 1.0 representing a binary classification model's ability to separate positive classes from negative classes. The closer the AUC is to 1.0, the better the model's ability to separate classes from each other. For example, the following illustration shows a classifier model that separates positive classes (green ovals) from negative classes (purple rectangles) perfectly. This unrealistically perfect model has an AUC of 1.0: Conversely, the following illustration shows the results for a classifier model that generated random results.

Accurate in silico modeling of the antigen processing pathway is crucial to enable personalized epitope vaccine design for cancer. An important step of such pathway is the degradation of the vaccine into smaller peptides by the proteasome, some of which are going to be presented to T cells by the MHC complex. While predicting MHC-peptide presentation has received a lot of attention recently, proteasomal cleavage prediction remains a relatively unexplored area in light of recent advancesin high-throughput mass spectrometry-based MHC ligandomics. Moreover, as such experimental techniques do not allow to identify regions that cannot be cleaved, the latest predictors generate decoy negative samples and treat them as true negatives when training, even though some of them could actually be positives. In this work, we thus present a new predictor trained with an expanded dataset and the solid theoretical underpinning of positive-unlabeled learning, achieving a new state-of-the-art in proteasomal cleavage prediction. The improved predictive capabilities will in turn enable more precise vaccine development improving the efficacy of epitope-based vaccines. Pretrained models are available on GitHub