May/June 2012

New Technology to Detect, Diagnose AD

By Juliann Schaeffer
Aging Well
Vol. 5 No. 3 P. 8

As the older adult population in the United States continues to grow, so will the number of individuals diagnosed with dementia, specifically Alzheimer’s disease (AD).

According to 2000 US Census estimates, dementia prevalence in individuals between the ages of 65 and 74 was 7%, with rates rising to nearly 50% in those aged 85 and older. “The number of older adults in the United States has been gradually increasing, with numbers expected to peak in 2029 when all baby boomers will have reached the age of 65,” says Ellen Yi-Luen Do, PhD, a professor in the School of Industrial Design and the School of Interactive Computing at the Georgia Institute of Technology in Atlanta.

“More than 5.3 million Americans have been diagnosed with Alzheimer’s disease, the most common form of dementia,” she says, noting that number is projected to increase to 13 million by 2050.

In addition to the physical and emotional devastation an AD diagnosis delivers to patients and their families, it also comes with a high financial cost. Direct costs to care for the 5.4 million Americans currently living with AD is projected to total an estimated $200 billion in 2012, according to the Alzheimer’s Association. The organization also notes that without significant changes, those costs will likely balloon to $1.1 trillion (in today’s dollars) by 2050.

Researchers at the Georgia Institute of Technology aim to effect such a change. Much remains unknown about the causes and potential cures for AD, but detecting it early gives doctors the best chance of delaying its progression. To that end, Georgia Tech researchers are working to develop innovative technologies to detect and diagnose AD at an earlier stage. ClockReader, developed by Do, is one such tool.

What Is It?
ClockReader is a technology aimed at improving one of the most commonly used screening tests to detect elders’ cognitive impairment: the clock drawing test (CDT). The CDT asks patients to draw a clock that indicates a particular time. Do worked from the traditional pen-and-paper tool to make the screening more accessible and effective by creating a software program that allows patients to take the test on a tablet PC.

The system includes two main parts: the ClockReader application for patients to draw a clock as well as the ClockAnalyzer application for clinicians to make appropriate diagnoses.

“The CDT is a paper-and-pencil test focusing on visual-spatial, constructional, and higher-order cognitive abilities,” says Do, explaining that the test directly maps neural damage to the impairment of visual-spatial behavior, illustrated by patients’ difficulty with number position or reversal of the minute and hour hand proportions.

The paper-and-pencil version has a mean sensitivity and specificity of 85%, yet Do explains that multiple scoring schemes exist, allowing scoring accuracy to vary based on how a particular physician interprets the test results.

“Though a clinically validated task, the scoring criteria and procedure still contain some inherent subjectivity and uncertainty,” she says. “In the Freedman scoring method, clinical technicians analyze hand-drawn clocks manually to score a set of 13 criteria. The grading instructions are consistent across technicians, but scoring accuracy varies due to different interpretations of correctness. Furthermore, the results of the paper CDT are shown to the doctor and then filed away without longitudinal analysis.”

The new screening, taken on a tablet PC, is meant to eliminate such drawbacks. “The drawing module of ClockReader records the CDT, and the viewing module displays drawing and scoring results and thumbnails of longitudinal drawings,” Do says. “ClockReader also provides an animated drawing playback, and timestamps of each stroke in the numerals reveal the patients’ drawing strategies—for example, 12-3-6-9 or 1-2-3-4.”

The software program then calculates the criteria automatically using the same 13-point evaluation system. While Do says no test currently available can give a 100% definitive diagnosis of AD or related dementias, she notes that user tests with volunteers and technicians have indicated comparable performance and scoring-consistency results between the paper-and-pencil test and the ClockReader test.

In addition, she says ClockReader can provide some additional insight into patients’ cognitive abilities, data that couldn’t be captured with the previous version of the test. “Studies have shown that mild cognitive impairment subject groups exhibit longer reaction times and more attention errors than the normal subject groups in performing neuropsychological tests,” Do explains. “Pressure data might differentiate different types of mild cognitive impairment. In one early Parkinson patient’s case, the patient’s mild hand tremor was recorded in the ClockReader but absent from the paper-and-pencil CDT.”

She says such data provide additional insight into the nature of mild cognitive impairment and enable clinical researchers to ask new questions about the relationship between the CDT and cognitive abilities.

Do’s team is currently working on adding new functionalities to ClockReader to gain even more data “to enable users to analyze and compare pressure and drawing sequence data, customize scoring criteria, create infrastructure for longitudinal analysis, and interactive visualizations to explore and extract meaning from novel sources of behavioral data,” she says.

Patient Feedback
Do says most patients who have used the ClockReader technology had no problems with it and weren’t intimidated by the technology. “Participants from Emory Alzheimer’s Disease Research Center Honor Research Registry found the ClockReader easy to use,” Do says. “Several of them commented that the tablet format is not like a computer and they like having a pen to interact and asked about where they can buy the tablet computer and the software program.”

Do says several of participants also expressed interest in using the program at home, showing interest in self-administering the tests and monitoring their own or a loved one’s condition.

Goals for the Clinical Setting
While she’s excited about the potential for increasing AD screening rates and earlier diagnoses in the clinical setting, Do and her team are working to improve the technology’s accuracy and user friendliness for patients, clinicians, and caregivers. They’re currently focusing on patients at Emory University Alzheimer’s Disease Research Center and its residential communities, namely Wesley Woods Senior Living towers. “I anticipate larger-scale deployment to happen this year,” she says.

Although the technology may not yet be ready for a full rollout, Do has a clear picture of where she’d like ClockReader to end up. “Eventually, I’d like to see all primary care physicians and other healthcare providers have access to the ClockReader technology to facilitate the quick and accurate identification of patients with possible cognitive decline in their practices,” she says.

Since primary care clinics provide the majority of healthcare to older adults in the United States, Do says this is a great place to start detecting AD in more patients at an earlier stage. Noting that evidence shows dementia fails to be recognized in 25% to 90% of patients seen in primary care, Do says, “The ClockReader and ClockAnalyzer project could provide clinicians with an efficient, affordable strategy to quickly screen their patients for possible cognitive decline in a busy primary care practice.

“Early identification of dementia in primary care settings can improve healthcare outcomes through earlier treatment of dementia and related comorbidities, implementing interventions to address safety concerns as well as to provide early access to educational and support services for the patient and family,” she adds.

The patient interface for ClockReader is completed, and Do says her team is developing the clinic view and ClockAnalyzer for clinicians to conduct further analysis in addition to scoring. She also hopes to be able to offer a consumer view of the technology to allow patients to self-monitor their cognitive condition much like patients can do with their weight or blood pressure.

“This will revolutionize the screening of Alzheimer’s disease and related disorders, as currently most cognitive declines were not identified in early onset,” Do says. “Having the convenience of screening at home or in a shopping mall or community centers could bring more awareness about the disease and related issues and concerns.”

Possible Future Uses
Though ClockReader was developed to screen for dementia, Do says the technology can be useful in detecting other cognitive impairments, such as traumatic brain injury and stroke.

“We are exploring several opportunities in using the ClockReader, ClockAnalyzer, and ComplexFigureAnalyzer [software currently under development that is used to do additional analysis on ClockReader results] to serve as a neurobehavior assessment method for multiple sclerosis, stroke, depression, and epilepsy patients,” she says. “With enough data, we may be able to differentiate the drawing patterns of different types of cognitive impairments.”

According to Do, ClockReader could also prove helpful in spinal cord and brain injury rehabilitation—and might even help patients with autism. “We know that Shepherd Center for spinal cord injury and brain injury rehabilitation [also in Atlanta] uses the clock-drawing test for rehabilitation. People have not looked into using the clock-drawing test as a rehab device for Alzheimer’s disease and related disorders, but we would like to explore this,” she says.

— Juliann Schaeffer is an associate editor at Great Valley Publishing Company.