MOTIVATION: In predicting HIV therapy outcomes, a critical clinical question is whether using historical information can enhance predictive capabilities compared with current or latest available data analysis. This study analyses whether historical knowledge, which includes viral mutations detected in all genotypic tests before therapy, their temporal occurrence, and concomitant viral load measurements, can bring improvements. We introduce a method to weigh mutations, considering the previously enumerated factors and the reference mutation-drug Stanford resistance tables. We compare a model encompassing history (H) with one not using this information (NH). RESULTS: The H-model demonstrates superior discriminative ability, with a higher ROC-AUC score (76.34%) than the NH-model (74.98%). Wilcoxon test results confirm significant improvement of predictive accuracy for treatment outcomes through incorporating historical information. The increased performance of the H-model might be attributed to its consideration of latent HIV reservoirs, probably obtained when leveraging historical information. The findings emphasize the importance of temporal dynamics in acquiring mutations. However, our result also shows that prediction accuracy remains relatively high even when no historical information is available. AVAILABILITY AND IMPLEMENTATION: This analysis was conducted using the Euresist Integrated DataBase (EIDB). For further validation, we encourage reproducing this study with the latest release of the EIDB, which can be accessed upon request through the Euresist Network.

OBJECTIVE: Clinical data's confidential nature often limits the development of machine learning models in healthcare. Generative adversarial networks (GANs) can synthesise realistic datasets, but suffer from mode collapse, resulting in low diversity and bias towards majority demographics and common clinical practices. This work proposes an extension to the classic GAN framework that includes a variational autoencoder (VAE) and an external memory mechanism to overcome these limitations and generate synthetic data accurately describing imbalanced class distributions commonly found in clinical variables. METHODS: The proposed method generated a synthetic dataset related to antiretroviral therapy for human immunodeficiency virus (ART for HIV). We evaluated it based on five metrics: (1) accurately representing imbalanced class distribution; (2) the realism of the individual variables; (3) the realism among variables; (4) patient disclosure risk; and (5) the utility of the generated dataset for developing downstream machine learning models. RESULTS: The proposed method overcomes the issue of mode collapse and generates a synthetic dataset that accurately describes imbalanced class distributions commonly found in clinical variables. The generated data has a patient disclosure risk of 0.095%, lower than the 9% threshold stated by Health Canada and the European Medicines Agency, making it suitable for distribution to the research community with high security. The generated data also has high utility, indicating the potential of the proposed method to enable the development of downstream machine learning algorithms for healthcare applications using synthetic data. CONCLUSION: Our proposed extension to the classic GAN framework, which includes a VAE and an external memory mechanism, represents a promising approach towards generating synthetic data that accurately describe imbalanced class distributions commonly found in clinical variables. This method overcomes the limitations of GANs and creates more realistic datasets with higher patient cohort diversity, facilitating the development of downstream machine learning algorithms for healthcare applications.

The lack of interpretability remains a barrier to adopting deep neural networks across many safety-critical domains. Tree regularization was recently proposed to encourage a deep neural network's decisions to resemble those of a globally compact, axis-aligned decision tree. However, it is often unreasonable to expect a single tree to predict well across all possible inputs. In practice, doing so could lead to neither interpretable nor performant optima. To address this issue, we propose regional tree regularization – a method that encourages a deep model to be well-approximated by several separate decision trees specific to predefined regions of the input space. Across many datasets, including two healthcare applications, we show our approach delivers simpler explanations than other regularization schemes without compromising accuracy. Specifically, our regional regularizer finds many more “desirable” optima compared to global analogues.

The lack of interpretability remains a key barrier to the adoption of deep models in many applications. In this work, we explicitly regularize deep models so human users might step through the process behind their predictions in little time. Specifically, we train deep time-series models so their class-probability predictions have high accuracy while being closely modeled by decision trees with few nodes. Using intuitive toy examples as well as medical tasks for treating sepsis and HIV, we demonstrate that this new tree regularization yields models that are easier for humans to simulate than simpler L1 or L2 penalties without sacrificing predictive power.

SUMMARY: RegaDB is a free and open source data management and analysis environment for infectious diseases. RegaDB allows clinicians to store, manage and analyse patient data, including viral genetic sequences. Moreover, RegaDB provides researchers with a mechanism to collect data in a uniform format and offers them a canvas to make newly developed bioinformatics tools available to clinicians and virologists through a user friendly interface. AVAILABILITY AND IMPLEMENTATION: Source code, binaries and documentation are available on http://rega.kuleuven.be/cev/regadb. RegaDB is written in the Java programming language, using a web-service-oriented architecture.

MOTIVATION: Several mathematical models have been investigated for the description of viral dynamics in the human body: HIV-1 infection is a particular and interesting scenario, because the virus attacks cells of the immune system that have a role in the antibody production and its high mutation rate permits to escape both the immune response and, in some cases, the drug pressure. The viral genetic evolution is intrinsically a stochastic process, eventually driven by the drug pressure, dependent on the drug combinations and concentration: in this article the viral genotypic drug resistance onset is the main focus addressed. The theoretical basis is the modelling of HIV-1 population dynamics as a predator-prey system of differential equations with a time-dependent therapy efficacy term, while the viral genome mutation evolution follows a Poisson distribution. The instant probabilities of drug resistance are estimated by means of functions trained from in vitro phenotypes, with a roulette-wheel-based mechanisms of resistant selection. Simulations have been designed for treatments made of one and two drugs as well as for combination antiretroviral therapies. The effect of limited adherence to therapy was also analyzed. Sequential treatment change episodes were also exploited with the aim to evaluate optimal synoptic treatment scenarios. RESULTS: The stochastic predator-prey modelling usefully predicted long-term virologic outcomes of evolved HIV-1 strains for selected antiretroviral therapy combinations. For a set of widely used combination therapies, results were consistent with findings reported in literature and with estimates coming from analysis on a large retrospective data base (EuResist).

MOTIVATION: Optimizing HIV therapies is crucial since the virus rapidly develops mutations to evade drug pressure. Recent studies have shown that genotypic information might not be sufficient for the design of therapies and that other clinical and demographical factors may play a role in therapy failure. This study is designed to assess the improvement in prediction achieved when such information is taken into account. We use these factors to generate a prediction engine using a variety of machine learning methods and to determine which clinical conditions are most misleading in terms of predicting the outcome of a therapy. RESULTS: Three different machine learning techniques were used: generative-discriminative method, regression with derived evolutionary features, and regression with a mixture of effects. All three methods had similar performances with an area under the receiver operating characteristic curve (AUC) of 0.77. A set of three similar engines limited to genotypic information only achieved an AUC of 0.75. A straightforward combination of the three engines consistently improves the prediction, with significantly better prediction when the full set of features is employed. The combined engine improves on predictions obtained from an online state-of-the-art resistance interpretation system. Moreover, engines tend to disagree more on the outcome of failure therapies than regarding successful ones. Careful analysis of the differences between the engines revealed those mutations and drugs most closely associated with uncertainty of the therapy outcome. AVAILABILITY: The combined prediction engine will be available from July 2008, see http://engine.euresist.org.

Predicting the in-vivo effect of genotypic drug resistance of Human Immunodeficiency Virus type-1 (HIV-1) on response to antiretroviral therapies represents a major clinical issue. Different machine learning and feature selection methods are applied for the classification of treatment success, based on viral genotype, therapy and derived input features. The robustness of results is assessed through statistical validation. The procedures described are intended to be a general methodology in the challenging context of biology and medical science data mining.