Predicting post-discharge complications in cardiothoracic surgery: A clinical decision support system to optimize remote patient monitoring resources.

Cardiac surgery patients are highly prone to severe complications post-discharge. Close follow-up through remote patient monitoring can help detect adverse outcomes earlier or prevent them, closing the gap between hospital and home care. However, equipment is limited due to economic and human resource constraints. This issue raises the need for efficient risk estimation to provide clinicians with insights into the potential benefit of remote monitoring for each patient. Standard models, such as the EuroSCORE, predict the mortality risk before the surgery. While these are used and validated in real settings, the models lack information collected during or following the surgery, determinant to predict adverse outcomes occurring further in the future. This paper proposes a Clinical Decision Support System based on Machine Learning to estimate the risk of severe complications within 90 days following cardiothoracic surgery discharge, an innovative objective underexplored in the literature. Health records from a cardiothoracic surgery department regarding 5 045 patients (60.8% male) collected throughout ten years were used to train predictive models. Clinicians' insights contributed to improving data preparation and extending traditional pipeline optimization techniques, addressing medical Artificial Intelligence requirements. Two separate test sets were used to evaluate the generalizability, one derived from a patient-grouped 70/30 split and another including all surgeries from the last available year. The achieved Area Under the Receiver Operating Characteristic curve on these test sets was 69.5% and 65.3%, respectively. Also, additional testing was implemented to simulate a real-world use case considering the weekly distribution of remote patient monitoring resources post-discharge. Compared to the random resource allocation, the selection of patients with respect to the outputs of the proposed model was proven beneficial, as it led to a higher number of high-risk patients receiving remote monitoring equipment.

Exploring Regularization Methods for Domain Generalization in Accelerometer-Based Human Activity Recognition.

The study of Domain Generalization (DG) has gained considerable momentum in the Machine Learning (ML) field. Human Activity Recognition (HAR) inherently encompasses diverse domains (e.g., users, devices, or datasets), rendering it an ideal testbed for exploring Domain Generalization. Building upon recent work, this paper investigates the application of regularization methods to bridge the generalization gap between traditional models based on handcrafted features and deep neural networks. We apply various regularizers, including sparse training, Mixup, Distributionally Robust Optimization (DRO), and Sharpness-Aware Minimization (SAM), to deep learning models and assess their performance in Out-of-Distribution (OOD) settings across multiple domains using homogenized public datasets. Our results show that Mixup and SAM are the best-performing regularizers. However, they are unable to match the performance of models based on handcrafted features. This suggests that while regularization techniques can improve OOD robustness to some extent, handcrafted features remain superior for domain generalization in HAR tasks.

Quantifying the Displacement of Data Matrix Code Modules: A Comparative Study of Different Approximation Approaches for Predictive Maintenance of Drop-on-Demand Printing Systems.

Drop-on-demand printing using colloidal or pigmented inks is prone to the clogging of printing nozzles, which can lead to positional deviations and inconsistently printed patterns (e.g., data matrix codes, DMCs). However, if such deviations are detected early, they can be useful for determining the state of the print head and planning maintenance operations prior to reaching a printing state where the printed DMCs are unreadable. To realize this predictive maintenance approach, it is necessary to accurately quantify the positional deviation of individually printed dots from the actual target position. Here, we present a comparison of different methods based on affinity transformations and clustering algorithms for calculating the target position from the printed positions and, subsequently, the deviation of both for complete DMCs. Hence, our method focuses on the evaluation of the print quality, not on the decoding of DMCs. We compare our results to a state-of-the-art decoding algorithm, adopted to return the target grid positions, and find that we can determine the occurring deviations with significantly higher accuracy, especially when the printed DMCs are of low quality. The results enable the development of decision systems for predictive maintenance and subsequently the optimization of printing systems.

Comparing Handcrafted Features and Deep Neural Representations for Domain Generalization in Human Activity Recognition.

Human Activity Recognition (HAR) has been studied extensively, yet current approaches are not capable of generalizing across different domains (i.e., subjects, devices, or datasets) with acceptable performance. This lack of generalization hinders the applicability of these models in real-world environments. As deep neural networks are becoming increasingly popular in recent work, there is a need for an explicit comparison between handcrafted and deep representations in Out-of-Distribution (OOD) settings. This paper compares both approaches in multiple domains using homogenized public datasets. First, we compare several metrics to validate three different OOD settings. In our main experiments, we then verify that even though deep learning initially outperforms models with handcrafted features, the situation is reversed as the distance from the training distribution increases. These findings support the hypothesis that handcrafted features may generalize better across specific domains.

Integrating Domain Knowledge into Deep Learning for Skin Lesion Risk Prioritization to Assist Teledermatology Referral.

Teledermatology has developed rapidly in recent years and is nowadays an essential tool for early diagnosis. In this work, we aim to improve existing Teledermatology processes for skin lesion diagnosis by developing a deep learning approach for risk prioritization with a dataset of retrospective data from referral requests of the Portuguese National Health System. Given the high complexity of this task, we propose a new prioritization pipeline guided and inspired by domain knowledge. We explored automatic lesion segmentation and tested different learning schemes, namely hierarchical classification and curriculum learning approaches, optionally including additional patient metadata. The final priority level prediction can then be obtained by combining predicted diagnosis and a baseline priority level accounting for explicit expert knowledge. In both the differential diagnosis and prioritization branches, lesion segmentation with 30% tolerance for contextual information was shown to improve classification when compared with a flat baseline model trained on original images; furthermore, the addition of patient information was not beneficial for most experiments. Curriculum learning delivered better results than a flat or hierarchical approach. The combination of diagnosis information and a knowledge map, created in collaboration with dermatologists, together with the priority achieved interesting results (best macro F1 of 43.93% for a validated test set), paving the way for new data-centric and knowledge-driven approaches.

An Infrastructure-Free Magnetic-Based Indoor Positioning System with Deep Learning.

Infrastructure-free Indoor Positioning Systems (IPS) are becoming popular due to their scalability and a wide range of applications. Such systems often rely on deployed Wi-Fi networks. However, their usability may be compromised, either due to scanning restrictions from recent Android versions or the proliferation of 5G technology. This raises the need for new infrastructure-free IPS independent of Wi-Fi networks. In this paper, we propose the use of magnetic field data for IPS, through Deep Neural Networks (DNN). Firstly, a dataset of human indoor trajectories was collected with different smartphones. Afterwards, a magnetic fingerprint was constructed and relevant features were extracted to train a DNN that returns a probability map of a user's location. Finally, two postprocessing methods were applied to obtain the most probable location regions. We asserted the performance of our solution against a test dataset, which produced a Success Rate of around 80%. We believe that these results are competitive for an IPS based on a single sensing source. Moreover, the magnetic field can be used as an additional information layer to increase the robustness and redundancy of current multi-source IPS.

Screening diabetes mellitus 2 based on electronic health records using temporal features.

The prevalence of type 2 diabetes mellitus is increasing worldwide. Current methods of treating diabetes remain inadequate, and therefore, prevention with screening methods is the most appropriate process to reduce the burden of diabetes and its complications. We propose a new prognostic approach for type 2 diabetes mellitus based on electronic health records without using the current invasive techniques that are related to the disease (e.g. glucose level or glycated hemoglobin (HbA1c)). Our methodology is based on machine learning frameworks with data enrichment using temporal features. As as result our predictive model achieved an area under the receiver operating characteristics curve with a random forest classifier of 84.22 percent when including data information from 2009 to 2011 to predict diabetic patients in 2012, 83.19 percent when including temporal features, and 83.72 percent after applying temporal features and feature selection. We conclude that he pathology prediction is possible and efficient using the patient's progression information over the years and without using the invasive techniques that are currently used for type 2 diabetes mellitus classification.

Prognostic models based on patient snapshots and time windows: Predicting disease progression to assisted ventilation in Amyotrophic Lateral Sclerosis.

Amyotrophic Lateral Sclerosis (ALS) is a devastating disease and the most common neurodegenerative disorder of young adults. ALS patients present a rapidly progressive motor weakness. This usually leads to death in a few years by respiratory failure. The correct prediction of respiratory insufficiency is thus key for patient management. In this context, we propose an innovative approach for prognostic prediction based on patient snapshots and time windows. We first cluster temporally-related tests to obtain snapshots of the patient's condition at a given time (patient snapshots). Then we use the snapshots to predict the probability of an ALS patient to require assisted ventilation after k days from the time of clinical evaluation (time window). This probability is based on the patient's current condition, evaluated using clinical features, including functional impairment assessments and a complete set of respiratory tests. The prognostic models include three temporal windows allowing to perform short, medium and long term prognosis regarding progression to assisted ventilation. Experimental results show an area under the receiver operating characteristics curve (AUC) in the test set of approximately 79% for time windows of 90, 180 and 365 days. Creating patient snapshots using hierarchical clustering with constraints outperforms the state of the art, and the proposed prognostic model becomes the first non population-based approach for prognostic prediction in ALS. The results are promising and should enhance the current clinical practice, largely supported by non-standardized tests and clinicians' experience.

Integrative biomarker discovery in neurodegenerative diseases.

Data mining has been widely applied in biomarker discovery resulting in significant findings of different clinical and biological biomarkers. With developments in technology, from genomics to proteomics analysis, a deluge of data has become available, as well as standardized data repositories. Nonetheless, researchers are still facing important challenges in analyzing the data, especially when considering the complexity of pathways involved in biological processes and diseases. Data from single sources appear unable to explain complex processes, such as those involved in brain-related disorders, including Alzheimer's disease, Parkinson's disease and amyotrophic lateral sclerosis, thus raising the need for a more comprehensive perspective. A possible solution relies on data and model integration, where several data types are combined to provide complementary views. This in turn can result in the discovery of previously unknown biomarkers by unraveling otherwise hidden relationships between data from different sources, and/or validate such composite biomarkers in more powerful predictive models.

Towards a classification approach using meta-biclustering: impact of discretization in the analysis of expression time series.

Biclustering has been recognized as a remarkably effective method for discovering local temporal expression patterns and unraveling potential regulatory mechanisms, essential to understanding complex biomedical processes, such as disease progression and drug response. In this work, we propose a classification approach based on meta-biclusters (a set of similar biclusters) applied to prognostic prediction. We use real clinical expression time series to predict the response of patients with multiple sclerosis to treatment with Interferon-ß. As compared to previous approaches, the main advantages of this strategy are the interpretability of the results and the reduction of data dimensionality, due to biclustering. This would allow the identification of the genes and time points which are most promising for explaining different types of response profiles, according to clinical knowledge. We assess the impact of different unsupervised and supervised discretization techniques on the classification accuracy. The experimental results show that, in many cases, the use of these discretization methods improves the classification accuracy, as compared to the use of the original features.

Prognostic prediction through biclustering-based classification of clinical gene expression time series.

The constant drive towards a more personalized medicine led to an increasing interest in temporal gene expression analyzes. It is now broadly accepted that considering a temporal perpective represents a great advantage to better understand disease progression and treatment results at a molecular level. In this context, biclustering algorithms emerged as an important tool to discover local expression patterns in biomedical applications, and CCC-Biclustering arose as an efficient algorithm relying on the temporal nature of data to identify all maximal temporal patterns in gene expression time series. In this work, CCC-Biclustering was integrated in new biclustering-based classifiers for prognostic prediction. As case study we analyzed multiple gene expression time series in order to classify the response of Multiple Sclerosis patients to the standard treatment with Interferon-ß, to which nearly half of the patients reveal a negative response. In this scenario, using an effective predictive model of a patient's response would avoid useless and possibly harmful therapies for the non-responder group. The results revealed interesting potentialities to be further explored in classification problems involving other (clinical) time series.