The system tends to remain relatively vague, leading to high inter-rater reliability since it depends on a clinician’s judgement for some generally described cases. However, it also comes with some downsides: Therefore, we selected the Shulman scoring system to score drawn clocks in this study.
Mental status clock drawing professional#
During active professional training for physicians after approval of antidementive drugs like donepezil and memantine in the 90ies until 2012, when the drugs were generic available, usually all treating physicians were trained and provided with the Shulman scheme by pharmaceutical companies, since regulation required regular assessment of treatment efficacy. The Shulman method 15, which provides valuable diagnostic information, is reliable, easy, valid and widely known and accepted in Germany. Several kinds of scoring systems are commonly used, such as MiniCog, Libon, Manos, Rouleau, Royall, and Shulman 10, 11, 12, 13, 14, 15. The result can then be screened and scored according to scoring systems. Within this test, an individual has to manually complete a clock on a piece of paper by drawing the numbers 1 to 12 as well as the clock hands for a certain time setting into a preprinted empty circle. The CDT is a brief, rapid and inexpensive paper-and-pencil cognitive screening tool also used to test for Parkinson or screening other cognitive changes 9. Note that it includes scoring the state of both, healthy and sick individuals.Ī test that is commonly used for screening and scoring dementia in the clinical routine is the so called ‘Clock-drawing test’ (CDT) 7, 8.
The score quantifies the progress of the disease and highly correlates to the patient’s mental state. ‘Scoring’ implies giving scores to a performed dementia test. ‘Screening’ is used to distinguish between sick and healthy individuals without differentiating between the levels of dementia or the state of mental healthiness. In this work, two different terms are used for diagnosing dementia. Furthermore, the method has to be available in a more digital manner such that more people can access it from anywhere in the world. Therefore, it is very important to develop an optimized diagnosing scheme, which can help to detect these syndromes as early as possible. Furthermore, around 60% of affected people live in low- and middle-income countries which sometimes have bad access to healthcare and limited money that can be spent for medical assistance or medication.Īs of today, there is no cure for these kinds of syndromes, but early detection and subsequent active management can dramatically improve patients’ and caregivers’ life qualities 3, 4, 5, 6. Based on US numbers, one in 10 people of age 65 and older lives with Alzheimer’s dementia and the total number of cases are projected to more than double by 2050 from 5.8 million to 13.8 million 2. According to the World Health Organization, around 50 million people currently have dementia and 10 million new cases are diagnosed each year 1. Due to the digital format, the paper-based test can be simply scanned by using a mobile device and then be evaluated also in areas where there is a staff shortage or where no clinical experts are available.ĭementia is one of the most common neurological syndromes in the world, in which the memory and the ability to perform daily life activities deteriorates and gets worse.
We achieved accuracies of 96.65% for screening and up to 98.54% for scoring, overcoming the reported state-of-the-art as well as human accuracies. The outcome of our work is a standardized and digital estimation of the dementia screening result and severity level for an individual. To deal with the limited amount of data, which also included several dementia types, we used optimization strategies for training the neural network. The dataset consisted of 1315 individuals. Our study includes a comparison of three modern architectures: VGG16, ResNet-152, and DenseNet-121. To overcome these issues, we present an automatic assessment of the widely used paper-based clock-drawing test by means of deep neural networks. This technique can introduce errors and has high inter-rater variability. Usually, diagnoses are made based on paper-and-pencil tests and scored depending on personal judgments of experts. Dementia is one of the most common neurological syndromes in the world.
0 kommentar(er)
