Analyzing entropy features in time-series data for pattern recognition in neurological conditions.

In the field of medical diagnosis and patient monitoring, effective pattern recognition in neurological time-series data is essential. Traditional methods predominantly based on statistical or probabilistic learning and inference often struggle with multivariate, multi-source, state-varying, and noisy data while also posing privacy risks due to excessive information collection and modeling. Furthermore, these methods often overlook critical statistical information, such as the distribution of data points and inherent uncertainties. To address these challenges, we introduce an information theory-based pipeline that leverages specialized features to identify patterns in neurological time-series data while minimizing privacy risks. We incorporate various entropy methods based on the characteristics of different scenarios and entropy. For stochastic state transition applications, we incorporate Shannon's entropy, entropy rates, entropy production, and the von Neumann entropy of Markov chains. When state modeling is impractical, we select and employ approximate entropy, increment entropy, dispersion entropy, phase entropy, and slope entropy. The pipeline's effectiveness and scalability are demonstrated through pattern analysis in a dementia care dataset and also an epileptic and a myocardial infarction dataset. The results indicate that our information theory-based pipeline can achieve average performance improvements across various models on the recall rate, F1 score, and accuracy by up to 13.08 percentage points, while enhancing inference efficiency by reducing the number of model parameters by an average of 3.10 times. Thus, our approach opens a promising avenue for improved, efficient, and critical statistical information-considered pattern recognition in medical time-series data.

Federated and distributed learning applications for electronic health records and structured medical data: a scoping review.

OBJECTIVES: Federated learning (FL) has gained popularity in clinical research in recent years to facilitate privacy-preserving collaboration. Structured data, one of the most prevalent forms of clinical data, has experienced significant growth in volume concurrently, notably with the widespread adoption of electronic health records in clinical practice. This review examines FL applications on structured medical data, identifies contemporary limitations, and discusses potential innovations. MATERIALS AND METHODS: We searched 5 databases, SCOPUS, MEDLINE, Web of Science, Embase, and CINAHL, to identify articles that applied FL to structured medical data and reported results following the PRISMA guidelines. Each selected publication was evaluated from 3 primary perspectives, including data quality, modeling strategies, and FL frameworks. RESULTS: Out of the 1193 papers screened, 34 met the inclusion criteria, with each article consisting of one or more studies that used FL to handle structured clinical/medical data. Of these, 24 utilized data acquired from electronic health records, with clinical predictions and association studies being the most common clinical research tasks that FL was applied to. Only one article exclusively explored the vertical FL setting, while the remaining 33 explored the horizontal FL setting, with only 14 discussing comparisons between single-site (local) and FL (global) analysis. CONCLUSIONS: The existing FL applications on structured medical data lack sufficient evaluations of clinically meaningful benefits, particularly when compared to single-site analyses. Therefore, it is crucial for future FL applications to prioritize clinical motivations and develop designs and methodologies that can effectively support and aid clinical practice and research.

Developing a Prognostic Model to Predict Mortality in Patients with Acute Bacterial Meningitis.

Bacterial meningitis is one of the harmful and deadly infectious diseases, and any delay in its treatment will lead to death. In this paper, a prognostic model was developed to predict the risk of death amongst probable cases of bacterial meningitis. Our prognostic model was developed using a decision tree algorithm on the national meningitis registry of the Iranian Center for Disease and Prevention (ICDCP) containing 3,923 records of meningitis suspected cases in 2018-2019. The most important features have been selected for the model construction. This model can predict the mortality risk for the meningitis probable cases with 78% accuracy, 84% sensitivity, and 73% specificity. The identified variables in prognosis the death included age and CSF protein level. CSF protein level (mg/dl) <= 65 versus > 65 provided the first branch of our decision tree. The highest mortality risk (85.8%) was seen in the patients >65 CSF protein level with 30 years < of age. For the patients <=30 year of age with CSF protein level >137 (mg/dl), the mortality risk was 60%. The prognostic factors identified in the present study draw the attention of clinicians to provide early specific measures, such as the admission of patients with a higher risk of death to intensive care units (ICU). It could also provide a helpful risk score tool in decision-making in the early phases of admission in pandemics, decrease mortality rate and improve public health operations efficiently in infectious diseases.

Modelling and forecasting art movements with CGANs.

Conditional generative adversarial networks (CGANs) are a recent and popular method for generating samples from a probability distribution conditioned on latent information. The latent information often comes in the form of a discrete label from a small set. We propose a novel method for training CGANs which allows us to condition on a sequence of continuous latent distributions f (1), …, f (K). This training allows CGANs to generate samples from a sequence of distributions. We apply our method to paintings from a sequence of artistic movements, where each movement is considered to be its own distribution. Exploiting the temporal aspect of the data, a vector autoregressive (VAR) model is fitted to the means of the latent distributions that we learn, and used for one-step-ahead forecasting, to predict the latent distribution of a future art movement f (K+1). Realizations from this distribution can be used by the CGAN to generate 'future' paintings. In experiments, this novel methodology generates accurate predictions of the evolution of art. The training set consists of a large dataset of past paintings. While there is no agreement on exactly what current art period we find ourselves in, we test on plausible candidate sets of present art, and show that the mean distance to our predictions is small.

Separation of cancer cells using vortical microfluidic flows.

Label-free separation of viable cancer cells using vortical microfluidic flows has been introduced as a feasible cell collection method in oncological studies. Besides the clinical importance, the physics of particle interactions with the vortex that forms in a wall-confined geometry of a microchannel is a relatively new area of fluid dynamics. In our previous work [Haddadi and Di Carlo, J. Fluid. Mech. 811, 436-467 (2017)], we have introduced distinct aspects of inertial flow of dilute suspensions over cavities in a microchannel such as breakdown of the separatrix and formation of stable limit cycle orbits for finite size polystyrene particles. In this work, we extend our experiments to address the engineering-physics of cancer cell entrapment in microfluidic cavities. We begin by studying the effects of the channel width and device height on the morphology of the vortex, which has not been discussed in our previous work. The stable limit cycle orbits of finite size cancer cells are then presented. We demonstrate effects of the separatrix breakdown and the limit cycle formation on the operation of the cancer cell separation platform. By studying the flow of dilute cell suspensions over the cavities, we further develop the notion of the cavity capacity and the relative rate of cell accumulation as optimization criteria which connect the device geometry with the flow. Finally, we discuss the proper placement of multiple cavities inside a microchannel for improved cell entrapment.

Building accountability into the Internet of Things: the IoT Databox model.

This paper outlines the IoT Databox model as a means of making the Internet of Things (IoT) accountable to individuals. Accountability is a key to building consumer trust and is mandated by the European Union's general data protection regulation (GDPR). We focus here on the 'external' data subject accountability requirement specified by GDPR and how meeting this requirement turns on surfacing the invisible actions and interactions of connected devices and the social arrangements in which they are embedded. The IoT Databox model is proposed as an in principle means of enabling accountability and providing individuals with the mechanisms needed to build trust into the IoT.

Selfish-herd behaviour of sheep under threat.

Flocking is a striking example of collective behaviour that is found in insect swarms, fish schools and mammal herds. A major factor in the evolution of flocking behaviour is thought to be predation, whereby larger and/or more cohesive groups are better at detecting predators (as, for example, in the 'many eyes theory'), and diluting the effects of predators (as in the 'selfish-herd theory') than are individuals in smaller and/or dispersed groups. The former theory assumes that information (passively or actively transferred) can be disseminated more effectively in larger/cohesive groups, while the latter assumes that there are spatial benefits to individuals in a large group, since individuals can alter their spatial position relative to their group-mates and any potential predator, thus reducing their predation risk. We used global positioning system (GPS) data to characterise the response of a group of 'prey' animals (a flock of sheep) to an approaching 'predator' (a herding dog). Analyses of relative sheep movement trajectories showed that sheep exhibit a strong attraction towards the centre of the flock under threat, a pattern that we could re-create using a simple model. These results support the long-standing assertion that individuals can respond to potential danger by moving towards the centre of a fleeing group.

Data-loggers carried on a harness do not adversely affect sheep locomotion.

The use of externally fitted motion sensors to animal subjects has the potential for allowing researchers to investigate subtle changes in animal movement that may occur with the onset of specific diseases. However, it is crucial to consider whether or not the use of such technology has an effect on the variables measured. Here, we examine the effect of a body harness data logging device on the locomotive patterns of female Merino sheep, Ovis aries. We extracted locomotion variables typical of motion sensor data (stride frequency, stride length, gait type, speed, and limb velocity) from high-definition video collected under controlled conditions. We found no significant difference between the variables measured in the harnessed and unharnessed conditions. Overall, our experiment demonstrates that data-loggers carried on a harness do not adversely affect sheep locomotion, and extended periods of habituation post-instalment of devices should ensure consistency and accuracy of data in future experiments.