Connect with us

Technology

Machine learning helps predict poor glycemic control in type 2 diabetes

Published

on

Machine learning helps predict poor glycemic control in type 2 diabetes

The risk for poor glycemic control in patients with type 2 diabetes can be predicted by using machine learning methods, a new study from Finland finds.

The most important factors predicting glycemic control include prior glucose levels, duration of type 2 diabetes, and the patient’s existing anti-diabetic medicines.

Researchers in Finland examined glycemic control in patients with type 2 diabetes in North Karelia, Finland, over a period of six years. A total of 9,631 people with type 2 diabetes were selected for the study cohorts. 

Patients’ glycemic control was determined on the basis of long-term blood glucose, HbA1c. Three HbA1c trajectories were identified from the data, and based on these, patients were divided into two groups: patients with adequate glycemic control, and patients with inadequate glycemic control. 

Using machine learning methods, the researchers examined the association of patients’ baseline characteristics, clinical- and treatment-related factors and socio-economic status with glycemic control. The baseline characteristics included more than 200 different variables.

The results showed that by using data on the duration of type 2 diabetes, prior HbA1c levels, fasting blood glucose, existing anti-diabetic medicines and their number, it is possible to reliably identify patients with a persistent risk for hyperglycemia at any point of their disease. In other words, inadequate glycemic control can be predicted from data that is routinely collected as part of diabetes monitoring and management.

The primary objective of treatment in type 2 diabetes is to maintain good glycemic control in order to prevent complications associated with the disease. According to the Finnish Current Care Guidelines for Diabetes, glycemic control should be followed up annually, making it possible to monitor the long-term trajectory of the disease. 

Early identification of patients with poor glycemic control is of paramount importance in order to target treatment to those in need and to intensify it at the right time. Delayed intensification of treatment increases the risk of complications, which is also reflected in higher costs of care.

The study utilised data from the electronic patient information system of the Joint Municipal Authority for North Karelia Social and Health Services, Siun sote, from registers maintained by the Social Insurance Institution of Finland, as well as from Statistics Finland’s open postal code database, Paavo. 

The study was carried out in collaboration between the University of Eastern Finland and the University of Oulu, and it was funded by the Finnish Diabetes Association, the Strategic Research Council at the Academy of Finland, Kuopio University Hospital (VTR funding) and the HTx project funded by the EU Horizon 2020 programme.

Technology

UK body calls for more ageing research backing

Published

on

The British Society for Research on Ageing (BSRA) is calling for more public backing in the UK for research to help people stay healthier for longer, as an alternative to charities that support research on diseases.

The greatest risk factor for disease is ageing, but we have very little charitable support for research into how to slow ageing, the organisation warns.

Many diseases such as cancers and heart disease tragically shorten lives far too early, or like Alzheimer’s and arthritis, destroy quality of life for patients and carers. There is understandably huge public charitable support for more research. However, the greatest risk factor for those diseases, and even infectious diseases like COVID, is ageing.

Yet in comparison there is currently very little support for research to understand how we can slow ageing to prevent disease. This approach may be more productive in the long term to fight disease. Furthermore, keeping people healthier for longer, or avoiding chronic diseases all together, would be the most favourable outcome.

The UK population is ageing fast, putting pressure on the NHS and the economy. Despite this pressing problem all around us, there is no accessible way for people to support research into ageing in the UK. The BSRA aims to change that.

With a very small budget and almost completely run by volunteers, the BSRA has successfully funded several small research projects but progress needs to be accelerated. More funding is needed because it takes years to see the effects of ageing, so studies are long. Also ageing affects individuals in different ways, meaning that large numbers of people must be studied to make firm conclusions.

Therefore, there is an urgency to get studies funded and the BSRA has decided to launch an ambitious fundraising campaign to boost research into ageing. Initially, the Society aims to fund a series of one year research projects at the Masters degree level at universities across the UK and with plans to raise much more in the future to support longer and more ambitious projects that will impact the lives of the general public.

Chair of the BSRA, Prof David Weinkove from Durham University, says “The time is now to really get behind research into the biology of ageing. We have fantastic researchers across the country, but they are held back by a lack of funding. Evidence-based research is needed to understand how we people can stay healthier for longer, and to then we must make that knowledge available to as many people as possible”.

Dr Jed Lye says “This is a great opportunity for the public to help, for corporations to contribute, or philanthropists wanting a large impact with a relatively small donation; every £20,000 we raise can fund an entire year of research into ageing and longevity, and gets a budding scientist their research qualification.”

Continue Reading

Technology

Wearable device could provide early warning of Alzheimer’s

Published

on

App for monitoring Parkinson’s disease gets FDA clearance

Monitoring daily activity patterns using a wrist-worn device may detect early warning signs of Alzheimer’s disease, according to a new study led by researchers at the Johns Hopkins Bloomberg School of Public Health.

The researchers analysed movement data from wristwatch-like devices called actigraphs worn by 82 cognitively healthy older adults who were participants in a long-running study of aging. Some of the participants had detectable brain amyloid

buildup as measured by PET scan. Buildup of the protein amyloid beta in the brain is a key feature of Alzheimer’s disease.

Using a sensitive statistical technique, the researchers found significant differences between this “amyloid-positive” group and “amyloid-negative” participants in mean activity in certain afternoon periods and differences in variability of activity across days in a broader range of time windows.

The new study was published online February 21 in the journal SLEEP.

“We need to replicate these findings in larger studies, but it is interesting that we’ve now seen a similar difference between amyloid-positive and amyloid-negative older adults in two independent studies,” says Adam Spira, PhD, professor in the Department of Mental Health at the Bloomberg School.

The new study’s results partly confirm findings from an earlier study in a smaller sample, also led by Spira, and suggest that actigraphs someday could be a tool to help detect incipient Alzheimer’s disease before significant cognitive impairment sets in. Data from the prior study came from participants in the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s (A4) and the Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) studies.

For their new study, Spira and colleagues investigated the potential of actigraph-based monitoring in 82 community-dwelling individuals whose average age was about 76. Each participant had a PET scan to measure brain amyloid and wore an actigraph 24 hours per day for one week. Using a sensitive statistical technique called FOSR (function-on-scalar regression), the researchers found that the 25 amyloid-positive participants, compared to the 57 amyloid-negative participants, had higher mean activity during the early afternoon, 1:00 to 3:30 p.m., and less day-to-day variability in activity from 1:30 to 4:00 p.m. and 7:30 to 10:30 p.m.

In more conservative analyses, some of these time windows with differences were no longer statistically significant. Nonetheless, the higher afternoon activity and lower afternoon variability echoed the investigators’ prior findings.

Alzheimer’s disease, the leading cause of dementia, is estimated to affect more than six million older adults in the U.S. The Alzheimer’s disease process is still not fully understood but is characterised by amyloid plaques and tangles in the brain, which typically begin to accumulate a decade or two before Alzheimer’s is diagnosed.

The only approved treatments that may slow the disease course are those that target amyloid beta and reduce the plaques. Many researchers believe that such treatments can be much more effective if given earlier in the disease course—ideally, years before dementia becomes evident.

Abnormal patterns of sleep and waking activity have been studied as potential early indicators of Alzheimer’s. Alzheimer’s patients typically have abnormal sleep-wake rhythms, and prior studies have found evidence that amyloid accumulation may disrupt sleep-wake rhythms relatively early in the disease process. There is also evidence that sleep loss promotes amyloid accumulation, suggesting a “vicious circle.”

Such findings hint at the possibility that older adults might someday, among other measures, wear wristwatch-like devices that would automatically track and analyse their sleep and waking activity. Individuals with anomalous activity patterns could then consult their physicians for more in-depth Alzheimer’s screening.

The individuals in the new study were participants in a long-running study, the Baltimore Longitudinal Study of Aging, which is conducted by the Intramural Research Program of the National Institute on Aging (NIA), part of the National Institutes of Health (NIH). Several members of the NIA team were co-authors of the study.

Standard, non-FOSR statistical methods did not detect any significant differences in activity or sleep patterns, suggesting the methods may be less sensitive to amyloid deposition.

In the earlier actigraphy study, the researchers, using FOSR-based analyses in a different sample of 59 participants, found increases in mean activity in afternoon hours and differences in variability, including lower variability in the afternoon, among amyloid-positive participants.

The scientists don’t know why amyloid buildup would trigger differences in activity patterns during these particular times of day. They note that there is a well-known phenomenon among individuals with Alzheimer’s disease called “sundowning,” in which agitation increases in the afternoon and early evening.

“It’s conceivable that the higher afternoon activity we observed is a signal of ‘preclinical sundowning,’” Spira says.

“At the same time, it’s important to note that these findings represent averages among a small sample of older people over a short period of time. We can’t predict whether an individual will develop amyloid plaques based on the timing of their activity. So, it would be premature for older people to be concerned because their fitness trackers say they are particularly active in the afternoon, for example.”

He and his colleagues plan to do larger studies of this kind. They also hope to do longer-term studies to see if daily activity-pattern changes are associated not only with brain amyloid but also with actual cognitive decline.

Continue Reading

Technology

Study detects cognitive changes in older drivers using in-vehicle sensors

Published

on

The forward-facing camera is mounted under the rearview mirror and is used to record events external to the vehicle. Photo credit: Jinwoo Jang, Ph.D., FAU College of Engineering and Computer Science

A new, in-vehicle sensing system could provide the first step toward widespread, low-cost early warnings of cognitive change among older drivers in the US and elsewhere.

An estimated 4 to 8 million older adults with mild cognitive impairment are currently driving in the United States, and one-third of them will develop dementia within five years. Individuals with progressive dementias are eventually unable to drive safely, yet many remain unaware of their cognitive decline.

Currently, screening and evaluation services for driving can only test a small number of individuals with cognitive concerns, missing many who need to know if they require treatment.

Nursing, engineering and neuropsychology researchers at Florida Atlantic University are testing and evaluating a readily and rapidly available, unobtrusive in-vehicle sensing system they have developed.

In their study, published in the journal BMC Geriatrics, they systematically examine how this system could detect anomalous driving behaviour indicative of cognitive impairment.

Few studies have reported on the use of continuous, unobtrusive sensors and related monitoring devices for detecting subtle variability in the performance of highly complex everyday activities over time. This significant proportion of older drivers constitutes a previously unexplored opportunity to detect cognitive decline.

Ruth Tappen, Ed.D., principal investigator, senior author and the Christine E. Lynn Eminent Scholar and Professor, FAU Christine E. Lynn College of Nursing, said: “The neuropathologies of Alzheimer’s disease have been found in the brains of older drivers killed in motor vehicle accidents who did not even know they had the disease and had no apparent signs of it.

“The purpose of our study arose from the importance of identifying cognitive dysfunction as early and efficiently as possible. Sensor systems installed in older drivers’ vehicles may detect these changes and could generate early warnings of possible changes in cognition.”

The study uses a naturalistic longitudinal design to obtain continuous information on driving behaviour that is being compared with the results of extensive cognitive testing conducted every three months for three years. A driver facing camera, forward facing camera, and telematics unit are installed in the vehicle and data is downloaded every three months when the cognitive tests are administered.

Researchers gauge abnormal driving such as getting lost, ignoring traffic signals and signs, near-collision events, distraction and drowsiness, reaction time and braking patterns. They also look at travel patterns such as number of trips, miles driven, miles on the highway, miles during the night and daytime, and driving in severe weather.

How it works

The in-vehicle sensor network developed by FAU researchers in the College of Engineering and Computer Science, uses open-source hardware and software components to reduce the time, risks and costs associated with developing in-vehicle sensing units.

In-vehicle sensor systems are kept simple and compact by minimising complex wiring, limiting the size of the sensing units, and limiting the number of sensors in a vehicle to support the unobtrusiveness of in-vehicle sensors. Each in-vehicle sensor system is comprised of two distributed sensing units: one for telematics data and the other for video data.

Inertial measurement unit data is processed to determine hard braking, hard accelerations and hard turns and GPS data. It also includes a timestamp, latitude, longitude, altitude, course over ground and the number of communicating satellites.

The video unit has built-in artificial intelligence functions that analyse video in real-time. The driver-facing camera is mounted in the left corner of the windshield and is directed to the driver’s face to analyse his/her behaviour and facial expressions. The forward-facing camera is mounted under the rearview mirror and is used to record events external to the vehicle.

Driver-facing indices include face detection, eye detection (open or closed), yawning, distraction, smoking and mobile phone use. Behaviour indices include traffic sign detection (running a red light), object detection (pedestrian, cyclists, curbs, barriers or nearby vehicles), lane crossing, near-collision and pedestrian detection.

“These travel-pattern-related driver behaviour indices are known to be indicative of the changes in older drivers’ cognition and physical functions since they tend to incorporate deliberate avoidance strategies to compensate for age-related deficits,” said Tappen.

“Driver behaviour indices are evaluated for each driver and are summarised on a daily, weekly and monthly basis and are classified into four categories.”

A total of 460 study participants will be recruited from Broward and Palm Beach counties in Southeast Florida and are classified into three diagnostic groups: mild cognitive impairment, early dementia and unimpaired (normal). The Louis and Anne Green Memory and Wellness Center operated by FAU’s College of Nursing serves as the testing site for a clinical battery including assessments of cognition, functioning in daily activities and mood (depression), and an additional set of tests including executive function and attention.

Tappen adds: “The innovation of our research project lies in the unobtrusive, rapidly and readily available in-vehicle sensing and monitoring system built upon modern open-source hardware and software using existing techniques to develop and customise the components and configure them for this new purpose.”

The study is supported by a grant from the National Institute on Aging.

Continue Reading

Trending