WebBioBank: A new platform for integrating clinical forms and shared neurosignal analyses to support multi-centre studies in Parkinson’s Disease

Knowledge-aware multi-center clinical dataset adaptation: Problem, method, and application

Adaptable utilization of clinical data collected from multiple centers, prompted by the need to overcome the shifts between the dataset distributions, and exploit these different datasets for potential clinical applications, has received significant attention in recent years. In this study, we propose a novel approach to this task by infusing an external knowledge graph (KG) into multi-center clinical data mining. Specifically, we propose an adversarial learning model to capture shared patient feature representations from multi-center heterogeneous clinical datasets, and employ an external KG to enrich the semantics of the patient sample by providing both clinical center-specific and center-general knowledge features, which are trained with a graph convolutional autoencoder. We evaluate the proposed model on a real clinical dataset extracted from the general cardiology wards of a Chinese hospital and a well-known public clinical dataset (MIMIC III, pertaining to ICU clinical settings) for the task of predicting acute kidney injury in patients with heart failure. The achieved experimental results demonstrate the efficacy of our proposed model.

N-of-1 Trials for Closed-Loop Deep Brain Stimulation Devices

Closed-loop deep brain stimulation (DBS) devices hold great promise for treating various neurological and psychiatric conditions. Yet while these algorithmic-based devices provide personalized treatment to each patient, they also present uniquely individualized risks of physiological and psychological harms. These experimental devices are typically tested in randomized controlled trials, which may not be the optimum approach to identifying and assessing phenomenological harms they pose to patients. In this article, we contend that an N-of-1 trial design-which is being used ever more frequently to realize the goals of individualized, precision medicine-could provide beneficial phenomenological data about the potential risks of harm to properly inform the use of closed-loop DBS devices. Data from N-of-1 trials may provide patients, as well as their families and other caregivers, with better information on which to base informed choices about pursuing this type of treatment option.

Deep brain stimulation in Parkinson's disease: A multicentric, long-term, observational pilot study

BACKGROUND

The impact of deep brain stimulation (DBS) on cognitive and urinary disorders, falls, and eventually hospitalizations and mortality in Parkinson's disease (PD) is still debated.

OBJECTIVE

We compared the rates of dementia, mild cognitive impairment (MCI), urinary incontinence, nocturia, falls, hospitalizations, and mortality in a cohort of PD patients undergoing DBS with a cohort of medically-treated patients chosen as controls.

METHODS

We conducted a retrospective pilot study in six Italian DBS centers. 91 PD patients receiving DBS and 91 age- and gender-matched controls receiving the best medical treatment alone with a minimum follow-up of one year were enrolled. Clinical data were collected from baseline to the last follow-up visit using an ad-hoc developed web-based system.

RESULTS

The risk of dementia was similar in the two groups while patients in the surgical cohort had lower rates of MCI, urinary incontinence, nocturia, and falls. In contrast, the risk of hospital admissions related to PD was higher in the surgical cohort. However, when excluding hospitalizations related to DBS surgery, the difference between the two cohorts was not significant. The surgical cohort had a lower number of hospitalizations not related to PD. The risk of death was similar in the two groups.

CONCLUSION

Despite a higher risk of hospitalization, patients receiving DBS had a lower rate of MCI, urinary incontinence, nocturia and falls, without evidence of an increased risk of dementia and mortality. Although these findings need to be confirmed in prospective studies, they seem to suggest that DBS may play a significant role in the management of non-motor symptoms and common complications of advanced PD.

Personally Collected Health Data for Precision Medicine and Longitudinal Research

Health data autonomously collected by users are presently considered as largely beneficial for wellness, prevention, disease management, as well as clinical research, especially when longitudinal, chronic, home-based monitoring is needed. However, data quality and reliability are the main barriers to overcome, in order to exploit such potential. To this end, we designed, implemented, and tested a system to integrate patient-generated personally collected health data into the clinical research data workflow, using a standards-based architecture that ensures the fulfillment of the major requirements for digital data in clinical studies. The system was tested in a clinical investigation for the optimization of deep brain stimulation (DBS) therapy in patients with Parkinson's disease that required both the collection of patient-generated data and of clinical and neurophysiological data. The validation showed that the implemented system was able to provide a reliable solution for including the patient as direct digital data source, ensuring reliability, integrity, security, attributability, and auditability of data. These results suggest that personally collected health data can be used as a reliable data source in longitudinal clinical research, thus improving holistic patient's personal assessment and monitoring.

Advances in closed-loop deep brain stimulation devices

BACKGROUND

Millions of patients around the world are affected by neurological and psychiatric disorders. Deep brain stimulation (DBS) is a device-based therapy that could have fewer side-effects and higher efficiencies in drug-resistant patients compared to other therapeutic options such as pharmacological approaches. Thus far, several efforts have been made to incorporate a feedback loop into DBS devices to make them operate in a closed-loop manner.

METHODS

This paper presents a comprehensive investigation into the existing research-based and commercial closed-loop DBS devices. It describes a brief history of closed-loop DBS techniques, biomarkers and algorithms used for closing the feedback loop, components of the current research-based and commercial closed-loop DBS devices, and advancements and challenges in this field of research. This review also includes a comparison of the closed-loop DBS devices and provides the future directions of this area of research.

RESULTS

Although we are in the early stages of the closed-loop DBS approach, there have been fruitful efforts in design and development of closed-loop DBS devices. To date, only one commercial closed-loop DBS device has been manufactured. However, this system does not have an intelligent and patient dependent control algorithm. A closed-loop DBS device requires a control algorithm to learn and optimize the stimulation parameters according to the brain clinical state.

CONCLUSIONS

The promising clinical effects of open-loop DBS have been demonstrated, indicating DBS as a pioneer technology and treatment option to serve neurological patients. However, like other commercial devices, DBS needs to be automated and modernized.

Web-based telemonitoring and delivery of caregiver support for patients with Parkinson disease after deep brain stimulation: protocol

Background

The increasing number of patients, the high costs of management, and the chronic progress of the disease that prevents patients from performing even simple daily activities make Parkinson disease (PD) a complex pathology with a high impact on society. In particular, patients implanted with deep brain stimulation (DBS) electrodes face a highly fragile stabilization period, requiring specific support at home. However, DBS patients are followed usually by untrained personnel (caregivers or family), without specific care pathways and supporting systems.

Objective

This projects aims to (1) create a reference consensus guideline and a shared requirements set for the homecare and monitoring of DBS patients, (2) define a set of biomarkers that provides alarms to caregivers for continuous home monitoring, and (3) implement an information system architecture allowing communication between health care professionals and caregivers and improving the quality of care for DBS patients.

Methods

The definitions of the consensus care pathway and of caregiver needs will be obtained by analyzing the current practices for patient follow-up through focus groups and structured interviews involving health care professionals, patients, and caregivers. The results of this analysis will be represented in a formal graphical model of the process of DBS patient care at home. To define the neurophysiological biomarkers to be used to raise alarms during the monitoring process, neurosignals will be acquired from DBS electrodes through a new experimental system that records while DBS is turned ON and transmits signals by radiofrequency. Motor, cognitive, and behavioral protocols will be used to study possible feedback/alarms to be provided by the system. Finally, a set of mobile apps to support the caregiver at home in managing and monitoring the patient will be developed and tested in the community of caregivers that participated in the focus groups. The set of developed apps will be connected to the already existing WebBioBank Web-based platform allowing health care professionals to manage patient electronic health records and neurophysiological signals. New modules in the WebBioBank platform will be implemented to allow integration and data exchange with mobile health apps.

Results

The results of this project will provide a novel approach to long-term evaluation of patients with chronic, severe conditions in the homecare environment, based on caregiver empowerment and tailored applications developed according to consensus care pathways established by clinicians.

Conclusions

The creation of a direct communication channel between health care professionals and caregivers can benefit large communities of patients and would represent a scalable experience in integrating data and information coming from a clinical setting to those in home monitoring.