Workflow adjustment of cognitive tests

Abstract

A method for adjusting a cognitive test routine in response to the detection of a patient's motor impairment during the execution of the cognitive test routine. In particular, a workflow for performing a neuropsychological evaluation including one or more cognitive tests is adjusted in response to the detection of motor impairment.

Description

Technical Field

[0001] The present invention relates to the field of cognitive testing, and in particular to the setting of a workflow for a series of cognitive tests.

Background Art

[0002] Many elderly people may develop movement disorders such as hand tremors, bradykinesia, or speech tremors. Parkinson's disease (PD), essential tremor (ET), and multiple sclerosis have a high prevalence of movement disorders. Sometimes, movement disorders are symptoms of underlying neurodegenerative diseases (e.g., PD), but in other cases, movement disorders may be benign and may be driven by drug therapy or other causes. Patients with cognitive impairment and / or neurodegenerative diseases are a major target group for neuropsychological assessment (e.g., on the Philips IntelliSpace Cognition (ISC) platform), so a significant number of subjects in neuropsychological assessment have some kind of movement disorder. Especially in the early stages, movement disorders are subtle and not reported by patients, and thus remain unknown to healthcare providers.

[0003] The most common movement disorder is tremor. This is an involuntary rhythmic muscle contraction that results in oscillatory movements in one or more parts of the body. Tremor often affects the hands, but can also affect the arms, head, vocal cords, torso, and legs. Tremor can be intermittent or constant. Tremor affecting the hand or arm can make it difficult to hold a pen and write or draw. Voice tremors can lead to a shaky voice that is difficult to understand.

Summary of the Invention

Problems to be Solved by the Invention

[0004] The execution of cognitive tests involving motor elements may be impaired by motor disorders, for example, because it may take patients longer to complete the tests or because patients may make more mistakes during the tests. For example, all tests on the ISC platform involve motor elements and may thus be affected by motor disorders. Tests such as drawing pictures or using a pen may be affected by hand tremors or bradykinesia. These include, for example, the Clock Drawing Test (CDT), Trail Making Test (TMT), and Rey Osterrieth Complex Figure Test (ROCFT). Tests that involve speech and may be affected by speech tremors include, for example, the Rey Auditory Verbal Learning Test (RAVLT) and the Controlled Oral Word Association Test (COWAT). Motor disorders may reduce the performance of these tests due to motor disorders rather than cognitive impairments.

[0005] Accordingly, the inventors recognized that there is a benefit in systems and methods that can assist in automatically determining whether specific results of cognitive tests have been affected by motor disorders.

Means for Solving the Problem

[0006] The present invention is defined by the claims.

[0007] One aspect of the present invention is a computer-implemented method, the method comprising:

[0008] controlling a test device to perform a neuropsychological assessment including motor elements on a subject, the neuropsychological assessment including performing a plurality of cognitive tests according to an ordered schedule defined by a neuropsychological assessment workflow and processing individual test results according to processing operations defined by the neuropsychological assessment workflow; Steps for automatically detecting one or more movement disorders during the performance of a neuropsychological assessment, steps for generating recommendations for workflow changes in the neuropsychological assessment in response to the detection, and steps for generating an output indicating the workflow change recommendations.

[0009] An object according to an embodiment of the present invention is to provide a tool that can assist in the differential detection of movement disorders as symptoms of neurodegenerative diseases, different from cognitive impairments. In the proposed method, this is based on detecting movement disorders during the execution of cognitive tests and adjusting the test workflow to compensate. This adjustment may relate to the calculation of cognitive test scores and / or to the execution of the test.

[0010] The neuropsychological assessment can include a series of cognitive tests, for example, a series of IntelliSpace Cognition (ISC) cognitive tests.

[0011] In some embodiments, the method can further include implementing the workflow change recommendations by changing the workflow according to the recommendations.

[0012] In some embodiments, the processing operation includes combining scores obtained for a plurality of individual cognitive tests of a neuropsychological assessment into one or more cognitive domain scores according to cognitive domain categories pre-associated with each cognitive test.

[0013] In some embodiments, the cognitive domain scores include one or more scores among memory, working memory, processing speed, language processing, executive function, and visual-spatial processing.

[0014] The workflow adjustment recommendations can include recommendations for modifying one or more cognitive domain scores.

[0015] This is to ensure that the individual test scores and domain test scores are not overly affected by movement disorders.

[0016] In some embodiments, the processing operation includes combining a plurality of cognitive region scores into an overall cognitive score, and the workflow adjustment recommendation includes a recommendation for modifying the overall cognitive score.

[0017] Workflow adjustments can be made at different times for the tests to be performed.

[0018] In some embodiments, in response to detecting a motor impairment during the performance of a particular cognitive test, a user-perceivable output indicating a workflow adjustment recommendation is generated during the neuropsychological evaluation, and the workflow change recommendation includes stopping the current cognitive test in which the motor impairment was detected.

[0019] In some embodiments, in response to detecting a motor impairment during the neuropsychological evaluation, a user-perceivable output indicating a recommendation is generated during the evaluation, and the workflow change recommendation includes skipping subsequent tests scheduled according to a workflow involving the same motor elements as the cognitive test in which the motor impairment was detected.

[0020] In some embodiments, in response to detecting a motor impairment during the neuropsychological evaluation, an output indicating a recommendation is generated after all the tests scheduled according to the workflow have been performed, and the workflow change recommendation includes adjusting the score combination part of the processing operation to compensate for the scores obtained for any individual cognitive test in which the motor impairment was detected.

[0021] In some embodiments, the scores of any tests in which a motor impairment was detected are removed from the score combination calculation.

[0022] In some embodiments, workflow change recommendations are generated only in response to the determined severity of the detected motor impairment exceeding a predetermined threshold and not otherwise.

[0023] In some embodiments, the workflow further includes generating a clinical report showing the results of the neuropsychological evaluation, and the workflow change recommendation includes adding an indication of any detected movement disorder to this report.

[0024] In some embodiments, detecting a movement disorder includes detecting the presence of any one or more of limb movement tremors, speech tremors, and bradykinesia. Of course, these examples are not exhaustive.

[0025] In some embodiments, the method further includes determining a quantitative result of the severity of the detected movement disorder.

[0026] In some embodiments, performing at least one of the cognitive tests includes obtaining a test data set that includes one or more data variables related to the performance of the test.

[0027] There are different ways to detect one or more movement disorders.

[0028] One approach is by spectral analysis.

[0029] In this regard, in some embodiments, detecting one or more movement disorders includes processing at least a subset of the test data set using spectral analysis to derive a spectral power spectrum, and detecting at least one movement disorder is performed based on an analysis of the spectral power spectrum. Spectral analysis can include Fourier analysis, and the spectral power spectrum is a Fourier power spectrum. As a further example, spectral analysis can include wavelet analysis.

[0030] In some embodiments, the movement disorder to be detected includes movement tremors. The aforementioned one or more data variables can include a data variable indicating the movement of a part of the subject's body as a function of time.

[0031] In some embodiments, the movement disorder to be detected can include vocal tremors. One or more of the aforementioned data variables can include a data variable that represents the audio acoustic data of the subject's speech input as a function of time.

[0032] In some embodiments, the movement disorder is determined based on detecting a deviation between the spectral power spectrum derived for the subject and a pre - memorized normal spectral power spectrum for a healthy control group.

[0033] In some embodiments, the severity of the tremor is quantified based on the derived spectral power spectrum.

[0034] In some embodiments, when the performance of at least one of the cognitive tests involves obtaining a test data set that includes a plurality of data variables related to the performance of the test, detecting one or more movement disorders includes applying a Bayesian model to the plurality of data variables collected for the test, and the Bayesian model is configured to output a probability score indicating the likelihood of the presence of a given movement disorder. This provides a multimodal detection of movement disorders.

[0035] In some embodiments, one or more Bayesian models are applied to generate respective probability scores for a plurality of different movement disorders, the plurality of probability scores are combined into a combined probability score, and workflow change recommendations are generated only in response to the combined probability score exceeding a predetermined threshold.

[0036] The combination may include forming a weighted combination.

[0037] The weighting coefficients can optionally be automatically estimated based on the variability of past and current performances.

[0038] In some embodiments, detecting a movement disorder includes distinctively identifying the type of the detected movement disorder, and the workflow change recommendation is configured according to the type of the detected cognitive disorder.

[0039] In some embodiments, the method further includes determining a risk prediction of cognitive decline based on a neuropsychological assessment.

[0040] In some embodiments, performing at least one of the cognitive tests includes obtaining a test data set including one or more data variables related to the performance of the test, and the method includes obtaining a comparator data set including the same data variables as those collected during the at least one cognitive test, but further including a calibration phase in which the at least one cognitive test is not performed.

[0041] In some embodiments, the comparator data set is analyzed to detect the presence of the same movement disorder as detected during the performance of the cognitive test and to quantify its severity.

[0042] The concept of the present invention can also be implemented in software form.

[0043] Accordingly, another aspect of the present invention is a computer program product comprising code means configured to cause a processor to execute a method according to any example or embodiment outlined in this document, or according to any claim of this application, when executed by the processor.

[0044] The present invention can also be implemented in hardware form. Accordingly, another aspect of the present invention is a system.

[0045] The system includes an input / output unit and one or more processors adapted, and the processor(s) The test device is controlled to perform a neuropsychological evaluation including motor elements on a subject via a control signal communicated through an input / output unit, and the neuropsychological evaluation includes performing a plurality of cognitive tests according to an ordered schedule defined by a neuropsychological evaluation workflow, and processing individual test results according to processing operations defined by the neuropsychological evaluation workflow. automatically detect one or more movement disorders during the performance of the neuropsychological evaluation generate neuropsychological evaluation workflow change recommendations in response to the detection, and be configured to generate an output indicating the workflow change recommendations.

[0046] The system can further include a memory storing a record of a standard neuropsychological workflow, the standard neuropsychological workflow defining an ordered schedule of cognitive tests to be performed on a subject and processing operations for processing individual test results.

[0047] The system can further include a test device for performing the neuropsychological evaluation on a subject.

[0048] These and other aspects of the invention will become apparent from the embodiments described below and will be described with reference thereto.

Brief Description of the Drawings

[0049] To better understand the present invention and to more clearly show how the present invention can be implemented, by way of mere example, reference is made to the accompanying drawings.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Best Mode for Carrying Out the Invention

[0050] The present invention will be described with reference to the drawings.

[0051] It should be understood that the detailed description and specific examples, while indicating exemplary embodiments of the apparatus, system and method, are for purposes of illustration only and are not intended to limit the scope of the present invention. These and other features, aspects, and advantages of the apparatus, system, and method of the present invention will be better understood from the following description, the appended claims, and the accompanying drawings. It should be understood that the drawings are merely schematic and are not drawn to scale. Also, it should be understood that the same reference numerals are used throughout the drawings to indicate the same or similar parts.

[0052] The present invention provides a method for adjusting a cognitive test routine that depends on the detection of a patient's movement disorder during the execution of a cognitive test. In particular, a workflow for performing a neuropsychological evaluation including one or more cognitive tests is adjusted in response to the detection of a movement disorder.

[0053] The digital neuropsychological assessment platform includes different cognitive tests. For example, the ISC platform provides a series of neuropsychological tests that evaluate different cognitive functions. Scores for individual neuropsychological tests are typically combined into cognitive domain scores according to the cognitive domain categories previously associated with each cognitive test. These can include, for example, memory, language processing, executive function, and / or visuospatial processing. FIG. 1 shows an overview of the cognitive tests available via the Philips IntelliSpace Cognition (Philips ISC) platform described above, and the contribution of each test to each of an exemplary set of cognitive domains. In the figure, the darkness of the gray indicates the magnitude of the contribution to the associated cognitive domain score, with the lightest gray indicating no contribution, and darker grays indicating that the contribution of a particular test to the cognitive domain is either smaller or larger in proportion to the darkness of the gray. FIG. 1 is a figure from the ISC product manual.

[0054] And in some cases, the cognitive domain scores are incorporated or combined into an overall cognitive score.

[0055] Readers are also referred to the literature from the Dent Neurologic Institute: Pinter, N.K., Bottoms, A., Mann, C., Ajtai, B., Pasternak, J., Harlock, S., & Fritz, J.V. (n.d.). Digital cognitive assessment in clinical practice utilizing Philips IntelliSpace Cognition: early clinical and operational experience. This literature provides a detailed consideration of the ISC tests and their applications in clinical practice.

[0056] Thus, in embodiments in which such composite scores are calculated, it will now be appreciated that the results of the individual cognitive tests can affect the cognitive domain scores and the overall cognitive score.

[0057] Thus, if the results of one or more cognitive tests are actually invalid for a particular patient due to the presence of a movement disorder (e.g., tremor) that distorts the results, the scores of these cognitive tests can unduly affect the cognitive domain scores and the overall cognitive score, and thus lead to an inaccurate assessment of cognitive function. This may be of particular concern when the neuropsychological assessment is a digital assessment and is performed remotely or in an unsupervised manner. Thus, it is important to know whether a movement disorder such as tremor is affecting the results of the cognitive tests. According to embodiments of the present invention, this information can be used to take corrective measures by adjusting the workflow of the neuropsychological assessment to accommodate the movement disorder, e.g., by adjusting the individual cognitive test scores, discounting the scores, removing a particular test from the domain and / or overall cognitive scores, aborting the execution of the test, or making various other workflow changes.

[0058] According to the present application, the inventors propose a method for automatically detecting a movement disorder during the execution of a series of digital cognitive tests (e.g., on the ISC platform already mentioned) and making workflow recommendations. When a movement disorder is detected, recommendations for workflow changes are made. As will become apparent through the examples and considerations presented below, workflow change recommendations will be understood as changes to the initial or standard implementation of the test for a particular digital neuropsychological assessment (e.g., the ISC platform), as well as the processing of individual test results and scores.

[0059] Figure 2 outlines the steps of an exemplary method according to one or more embodiments in a block diagram. The steps are summarized before being further described in the form of exemplary embodiments.

[0060] Method 10 includes controlling a test device to perform a neuropsychological evaluation including motor elements on a subject (12). The neuropsychological evaluation includes performing a plurality of cognitive tests according to an ordered schedule defined by a neuropsychological evaluation workflow. The neuropsychological evaluation further includes processing individual cognitive test results according to processing operations defined by the neuropsychological evaluation workflow.

[0061] Method 10 further includes automatically detecting one or more movement disorders during the execution of the neuropsychological evaluation (14). By way of example, the movement disorders that can be detected can include any one or more of limb movement tremors, speech tremors, and bradykinesia.

[0062] Method 10 further includes generating neuropsychological evaluation workflow change recommendations in response to the detection (16).

[0063] Method 10 further includes generating an output indicating the workflow change recommendation (18).

[0064] Output 18 can be a user-perceivable output, such that the final result of the method is simply a recommendation communicated to a user, such as a subject undergoing the test or a clinician supervising the administration of the test.

[0065] The output can be a data output. The data output can be communicated to one or more additional locations or addresses.

[0066] The method can further include implementing the workflow change recommendation by changing the workflow according to the recommendation.

[0067] As described above, the method can also be implemented in hardware form.

[0068] For purposes of illustration, FIG. 3 shows a schematic diagram of an exemplary system 30 according to one or more embodiments of the present invention.

[0069] The illustrated system includes a processing unit 32 having an input / output portion 34 and one or more processors 36.

[0070] The one or more processors 36 are configured to perform a method according to any example or embodiment described herein, or according to any claim of the present application. More explicitly, the one or more processors 36 may be adapted to perform the following steps: · Controlling the test device 42 to perform a neuropsychological evaluation on a subject, including motor elements, via a control signal communicated through the input / output portion 34, the neuropsychological evaluation including performing a plurality of cognitive tests according to an ordered schedule defined by a neuropsychological evaluation workflow, and processing individual test results according to processing operations defined by the neuropsychological evaluation workflow; · Automatically detecting one or more movement disorders during the execution of the neuropsychological evaluation; · Generating a neuropsychological evaluation workflow change recommendation in response to the detection; and, · Generating an output indicating the workflow change recommendation.

[0071] One aspect of the present invention is a system comprising only the processing unit 32 (or in another expression, only the input / output portion 34 and one or more processors 36).

[0072] Optionally, the system 30 may further include a memory 38 that stores a record of a standard neuropsychological workflow, the standard neuropsychological workflow defining an ordered schedule of cognitive tests to be performed on a subject and defining processing operations for processing individual test results.

[0073] As an option, the system can include a test device 42 for performing neuropsychological evaluations on a subject. For example, this can be a computer-based test device, i.e., it can include a computing device. It can be a dedicated computing console or terminal. It can include a mobile computing device such as a tablet computer, a smartphone, or a laptop. It can include a user interface. The user interface can include one or more sensory output devices for generating stimuli perceptible to the user. The user interface can include a user input device. This can include an electronic pen or pencil, and / or one or more pointer devices such as a touch screen, and / or a gesture capture device such as a handheld unit having an inertial measurement device integrated therein, or a camera / sensor-based gesture capture device.

[0074] In any embodiment, the neuropsychological evaluation can be a digital neuropsychological evaluation. A digital neuropsychological evaluation includes a neuropsychological evaluation that is performed via control of a user interface device to generate prompt instructions and measure test response data, i.e., a test that is electronically performed.

[0075] As described above, in a typical neuropsychological evaluation workflow, the scores of individual neuropsychological tests are combined into cognitive domain scores, e.g., memory, executive function. The cognitive domain scores can further be merged or combined into an overall cognitive score.

[0076] Thus, in some embodiments of the present invention, the processing operation part of the workflow includes combining the scores obtained for a plurality of individual cognitive tests of a neuropsychological evaluation into one or more cognitive domain scores according to the cognitive domain categories pre-associated with each cognitive test. This combining operation can be defined by a combining function that specifies, in a standard workflow, the weighting to be applied to the individual cognitive test scores in the combination.

[0077] As an example, the cognitive domain score can include any score for any one or more of memory, language processing, executive function, visual-spatial processing.

[0078] Workflow adjustment recommendations can, in some embodiments, include recommendations for correcting one or more cognitive domain scores. It can additionally or alternatively include changes to the weighting associated with one or more of the individual cognitive test scores in the combination that forms the cognitive domain score. It can additionally or alternatively include the exclusion of one or more cognitive test scores from the cognitive domain score calculation.

[0079] Furthermore, according to some embodiments, the processing operation part of the workflow may include combining a plurality of cognitive domain scores into an overall cognitive score. Here, in some embodiments, the workflow adjustment recommendations include recommendations for modifying the overall cognitive score.

[0080] There are different options regarding when and / or whether workflow change recommendations are made within the workflow of a neuropsychological evaluation.

[0081] In some embodiments, the recommendations can be made during the neuropsychological evaluation, for example, such that a test in which a movement disorder is detected is stopped.

[0082] For example, in some embodiments, in response to detecting a motor impairment during the performance of a particular cognitive test, a user-perceivable output indicating a workflow adjustment recommendation is generated during the neuropsychological evaluation, and the workflow change recommendation includes stopping the current cognitive test in which the motor impairment was detected.

[0083] As another example, the workflow change recommendation can be made during the evaluation such that, for example, subsequent tests that include the same motor element (i.e., drawing, writing, or speaking) are skipped from the neuropsychological evaluation.

[0084] For example, in some embodiments, in response to detecting a motor impairment during the neuropsychological evaluation, a user-perceivable output indicating a recommendation is generated during the evaluation, and the workflow change recommendation includes skipping subsequent tests scheduled according to a workflow that involves the same motor element as the cognitive test in which the motor impairment was detected.

[0085] According to any of the above examples, the user-perceivable output can be, for example, a visual output or an auditory output.

[0086] In some embodiments, the workflow change recommendation can be made after the evaluation, and as a result, the test in which the motor impairment was detected is removed from the calculation of the relevant cognitive domain test score and / or the overall cognitive score.

[0087] For example, in some embodiments, in response to detecting a motor impairment during the neuropsychological evaluation, the output indicating a recommendation is generated after all the tests scheduled according to the workflow have been performed, and the workflow change recommendation includes adjusting the score combination part of the processing operation to compensate for the score obtained for any individual cognitive test in which the motor impairment was detected. For example, the score of any test in which the motor impairment was detected may be removed from the score combination calculation.

[0088] In any of the above, it is optional that workflow change recommendations are generated only in response to the determined severity of the detected movement disorder exceeding a predetermined threshold and not otherwise. Options for how the severity can be determined are described below.

[0089] In some embodiments, the workflow can further include generating a clinical report that presents the results of a neuropsychological evaluation.

[0090] In some embodiments, the workflow recommendations can include tremor warnings or descriptions that may be added to the clinical report. This change to the workflow shortens the evaluation time and ensures that the outcome of the overall cognitive evaluation or specific elements is reliable and valid.

[0091] Regarding the detection of movement disorders, different options are possible. Generally, it is based on processing one or more data variables included in a test data set obtained as part of a cognitive test with a dedicated detection algorithm that detects the presence or absence of at least one specific movement disorder and preferably determines a measure of its severity.

[0092] One advantageous approach is to use spectral analysis of the cognitive test data set, such as Fourier analysis or wavelet analysis.

[0093] In this approach, one or more movement disorders are detected based on the processing of at least a subset of the acquired cognitive test data set using spectral analysis to derive a frequency domain power spectrum, such as a Fourier power spectrum, and the detection of at least one movement disorder is performed based on spectral power spectrum analysis. The cognitive test data set is a data set obtained as part of at least one normal performance of a cognitive test and includes one or more data variables related to the results of the test.

[0094] As a general principle, a given movement disorder can be detected and identified in a manner based on detecting a deviation between a derived spectral power spectrum for a subject and a pre - memorized normal spectral power spectrum for a healthy control group. A dedicated comparison algorithm can be provided for discriminatively detecting each of one or more different movement disorders based on applying different detection criteria to any detected deviation and / or detecting or extracting the deviation in different ways.

[0095] In some embodiments, the severity of a movement disorder can be quantified based on the derived spectral power spectrum.

[0096] To illustrate this concept, two exemplary implementations will be described with reference to FIGS. 4 and 5 respectively. It should be understood that not all features of each of these specific examples are essential to the concept of the present invention, and each is described to assist in understanding and to provide an example for illustrating the concept of the present invention.

[0097] The first example described with reference to FIG. 4 relates to the case where the movement disorder to be detected includes movement tremors. The cognitive test being performed requires the user to execute a drawing task via the movement of a pencil pointer device on the input means of a computing device. In the example of FIG. 4, the input device is a touch - screen display 52. The computing device is a tablet computer in the illustrated example.

[0098] A test data set is acquired that includes a data variable corresponding to the movement path of the pencil pointer device on screen 52.

[0099] Thus, this is an example of a more general case where the implementation of at least one of a series of cognitive tests includes the acquisition of a test data set that includes a data variable indicative of the movement of a part of the subject's body as a function of time.

[0100] Next, through the automatic analysis of tactile input provided via the stylus of the tablet, the quantification of motor tremors is determined according to the set of this embodiment. The analysis is schematically shown in FIG. 4. In this illustrated example, the test corresponds to the Trail Making Test (TMT) of the ISC platform described above.

[0101] This process involves a comparison of the Fourier spectrum derived from the captured tactile input (trace b) with the baseline spectrum for a healthy population (trace a).

[0102] The drawing trajectory captured by the input device 52 is processed as a function of time, i.e., to extract the x, y, and z position variables 62b as x, y, z time series. This can involve, for example, a normalization process that uses calibration data already obtained prior to the workflow, either via an instruction / test session or when the test subject is writing their name. This normalization can be performed in real time while the input data is being acquired, i.e., while the user is drawing on the screen 52, or it can be performed in batch form at a later stage.

[0103] The x, y, and z position variable data is processed to obtain the first-order time derivative data 64b of time x, y, and z. Next, a Fourier transform is applied to the first-order derivative signals 64b of x, y, z over a specified time window, from which the Fourier power spectrum 66a is derived. Alternatively, the Fourier transform can be applied directly to the position variable signal 62a.

[0104] Analysis including comparison or contrast with Fourier spectrum 66a against a healthy control group is performed. FIG. 4 schematically shows the intermediate steps of deriving the position trajectory data 62a, first derivative data 64a, and Fourier spectrum 66a of a healthy control group to further assist understanding and explanation. However, in reality, these steps are executed prior to the method of the present invention, the Fourier spectrum is pre-stored, and it is expected that preparations for comparison or contrast with the newly captured data spectrum 66b to be performed are complete. However, of course, it is also conceivable to perform Fourier analysis of the control data in real time.

[0105] The deviation of the Fourier power spectrum 66b derived from the test data set from the normal power spectrum 66a is determined over an equivalent time window. From this, the frequency peaks in the test spectrum 66b corresponding to tremors can be estimated. Then, the frequency band of each tremor, the amplitude of each tremor (e.g., from the height of each peak), and the number of tremors (e.g., from the number of peaks) can be detected and used to quantify the tremors.

[0106] In the illustrated example, distinct peaks can be detected in the test Fourier spectrum 66b. By quantifying this peak (e.g., spectral power or area under the curve) relative to the normal spectrum 66a, hand tremors, for example, can be detected.

[0107] The second example described with reference to FIG. 5 relates to the case where the movement disorder to be detected includes voice tremors. Detection and preferably quantification of voice tremors are determined via automatic analysis of voice input provided, for example, via the microphone of the tablet computer 52.

[0108] Therefore, this is an example of a more general case where the implementation of at least one of a series of cognitive tests involves obtaining a test data set that includes a data variable indicative of the audio acoustic data of the subject's speech input as a function of time. As an example, voice-based cognitive tests in the ISC platform include the Rey Auditory Verbal Learning Test (RAVLT) and the Controlled Oral Word Association Test (COWAT).

[0109] Row (a) corresponds to data of a healthy control group. Row (b) corresponds to data obtained for a subject during a cognitive test with voice tremor. The acoustic amplitude data 72b is obtained during the execution of the test and is performed in the form of wavelet analysis over a time window in which spectral analysis is defined for this, generating a power spectrum 74b in the frequency domain. Alternatively, Fourier analysis can be performed. Next, peaks in the power spectrum can be derived. Based on the deviation from the normal power spectrum 74a derived from the acoustic data 72a for the normal control group, dyskinesia can be detected and quantified. For example, the detected number of deviations (e.g., the number of peaks present in the test data spectrum that are not present in the normal spectrum) can be used to quantify voice tremor. Additionally or alternatively, the area under the curve of the deviation peaks detected in the test spectrum 74b can be used to quantify tremor and / or the height of such peaks.

[0110] Instead of a single data variable, the implementation of at least one of the cognitive tests involves obtaining a test data set that includes a plurality of data variables related to the performance of the test, and if the plurality of data variables of the test data set are included in the analysis, a more accurate and discriminatory detection of one or more dyskinesias is possible. This is known in the art as multimodal evaluation.

[0111] For example, referring to the examples described above with reference to FIGS. 4 and 5, additionally, information regarding the speed and changes in speed of the pen movement, the angle of the pen on the tablet and changes in the angle of the pen, the number and speed of direction changes, and the pressure and changes in pressure during drawing and / or writing can be monitored and acquired as part of the test dataset. And these data variables can be used for multimodal detection of movement disorders that affect the hand and / or arm.

[0112] The multimodal evaluation may be specific to a given cognitive test and may take into account the characteristics of a particular cognitive test. For example, in the Trail Making Test (TMT), speed changes are expected due to the visual distribution of the targets. The speed decreases as it approaches the target and increases as it moves away from the target. The speed is also affected by the distance between two consecutive targets. One solution is to define a perimeter around the target and exclude speed changes within this perimeter from the analysis. These metrics (speed, angle, direction change, pressure) can together be used for multimodal detection of tremors.

[0113] Based on multiple measured test variables, or parameters derived from them (e.g., from spectral transforms), a movement disorder "score" can be determined to represent the probability P(T|C) that a particular movement disorder, such as tremor, has occurred when both prior knowledge and the measured parameters of the test variables or the spectral transform of the tremor during the task are given.

[0114] The method of combining likelihood and prior knowledge follows Bayesian probability theory.

[0115] For example, as a general principle, when at least one implementation of a cognitive test involves obtaining a test data set that includes a plurality of data variables, the detection of one or more movement disorders in some embodiments can include applying a Bayesian model to the plurality of data variables or parameters derived therefrom, such as parameters of a spectral transform. The Bayesian model can be configured to output a probability score P(T|C) indicating the likelihood of the presence of a given movement disorder, as described above.

[0116] Here, further details are described by way of an illustrative embodiment where the movement disorder to be detected is tremor. However, it should be understood that the principles of this approach can be more broadly applied to the detection of any movement disorder.

[0117] In a Bayesian model, general inference is first decomposed into a "likelihood" probability distribution. This represents the probability of one or more cues (C) given a particular sensorimotor disorder data (M), denoted as P(C|M). Second, it is further decomposed into what is called the "prior probability". This represents the prior expected probability associated with the sensorimotor disorder data P(M). Combining the likelihood and the prior probability, the "a posteriori" probability P(M|C) represents the probability that a particular parameter in the sensorimotor disorder data (M) exists given the cue (C), and is obtained from Bayes' rule in the general form: P(T|C) ∝ P(C|T)P(T). If this score (probability) exceeds a given threshold (e.g., the population mean), a flag is set in the log file. Also, the (prior) conditions that were below the threshold are recorded in the log file.

[0118] The process for constructing and implementing an appropriate Bayesian model will be well known to those skilled in the art. For example, the Internet Encyclopedia article: https: / / en.wikipedia.org / wiki / Bayes_theorem may be referred to. Any textbook on Bayesian theory can also be used as a further reference.

[0119] In some embodiments, a composite multimodal tremor "score" can be determined to quantify the role of multiple combined impairments (e.g., hand tremor, voice tremor).

[0120] In this regard, as a general principle, in some embodiments, one or more Bayesian models are applied to generate respective probability scores for multiple different movement impairments, and the multiple probability scores are combined to form a combined probability score. In some embodiments, this quantitative value of the score can be used to determine whether a workflow change is needed and which workflow change should be made. In some embodiments, workflow change recommendations may be generated only in response to a combined probability score exceeding a predefined threshold.

[0121] As an example, a patient may exhibit multiple tremors, each with a specific likelihood and a prior state in which the tremor occurs. Combining the effects of multiple tremors can be valuable as multiple types of tremors may interfere with cognitive function in a cumulative manner. Weighting factors can be used to combine the multiple individual tremor scores. The weighting factors can be automatically estimated, for example, based on past and current performance variability (e.g., variability of the above-described parameters from the spectral power spectrum). For example, a composite tremor score can be calculated from the scores for individual tremors as follows: Weight 1 x Tremor 1 + Weight 2 x Tremor 2 = Composite Tremor

[0122] To summarize the concept, FIG. 6 outlines an exemplary process flow for an exemplary Bayesian approach for movement disorder detection.

[0123] This process includes, at a first stage 82, determining respective movement disorder probability scores for each of multiple different movement impairments, for example, in the manner already described above.

[0124] This method further includes, in a second stage 84, a multimodal processing operation that combines the plurality of individual probability scores obtained in stage 82 with respective weightings to form a composite movement disorder score (e.g., a composite tremor score in the example of FIG. 6).

[0125] This method further includes, in a third stage 86, a determination step of evaluating the composite score with reference to one or more predefined criteria for the algorithm to determine the presence or absence of a movement disorder. For example, in the example shown in FIG. 6, the composite score is compared with a predetermined threshold value.

[0126] This method further includes, in stage 88, generating a recommendation based on the determination step of stage 86. This may be a recommendation perceptible to the user and can be communicated to the patient and / or the neurologist performing the neuropsychological assessment. Different recommendations, such as those listed as (1), (2), and (3) in the example of FIG. 6, can be communicated to the patient and the neurologist respectively. For example, a recommendation to stop the test can be issued to the patient, and in addition, an update of the test result calculation, i.e., a recommendation for interpretation, can be communicated to the neurologist.

[0127] In some embodiments, the analysis applied in accordance with the present invention can distinguish different types of movement disorders from each other. For example, tremors can be distinguished from bradykinesia, and head tremors can be distinguished from laryngeal (voice) tremors. This is valuable because it is known that some types of tremors interfere more with cognitive function.

[0128] Thus, as a general principle, in some embodiments, detecting a movement disorder includes distinctively identifying the type of detected movement disorder, and it can be said that the workflow change recommendations are configured according to the type of detected cognitive disorder.

[0129] The detection of movement disorders can be used to signal further analysis or detection. For example, the risk of future cognitive impairment, or the progression of cognitive impairment, can be predicted at least in part based on the detected movement disorder and severity or other parameters (e.g., type). Thus, in some embodiments, the method can further include determining a risk prediction of cognitive decline based on a neuropsychological evaluation.

[0130] For example, studies have shown that tremors at the onset of Parkinson's disease (PD) are significant predictors of worse cognitive abilities. With respect to different types of tremors, studies have shown that PD patients with primary akinetic and rigid tremors, but not those with essential tremors, have decreased activity in the default mode network and smaller gray matter volumes compared to healthy controls (Karunanayaka, P.R., Lee, E.-Y., Lewis, M.M., Sen, S., Eslinger, P.J., Yang, Q.X., & Huang, X. (2017). Default mode network differences between rigidity- and tremor-predominant Parkinson's disease. Cortex, 81, 239-250). Decreased activity in the default mode network is associated with cognitive impairment in PD (ibid). Using various variables such as not only the magnitude and type of tremors but also cognitive dysfunction (corrected for motor dysfunction), the (further) risk of long-term cognitive decline in neurodegenerative diseases can be predicted.

[0131] In addition to quantifying the severity of movement disorders during the performance of a series of cognitive tests, a calibration phase can be implemented in which the analysis parameters are set to enable subsequent quantification of potential movement disorders (e.g., according to the detected data variables such as frequency, amplitude, intensity, asymmetry, etc.).

[0132] In particular, in some embodiments, when the performance of at least one of the cognitive tests involves obtaining a test data set that includes one or more data variables related to the performance of the test, the method can further include a calibration phase that includes obtaining a comparator data set that includes the same data variables as were collected during the at least one cognitive test, but without performing the at least one cognitive test.

[0133] For example, a dedicated calibration data set of data variables can be obtained while at rest, i.e., while no task is being performed. In many cases, such as in PD, tremors worsen at rest compared to when performing planned movements. When a movement plan is initiated and executed, the severity of the tremors may decrease. Therefore, it is interesting to compare the severity of tremors at rest (before test implementation) with the severity of tremors during the execution of a series of cognitive tests.

[0134] During the analysis of the cognitive test data set for detecting movement disorders, the comparator data set can be utilized to detect the presence of the same movement disorders detected during the implementation of the cognitive test and to quantify their severity.

[0135] As already emphasized, the examples in this specification tended to focus on tremors, but other types of movement disorders, such as bradykinesia or rigidity, can also be detected in a similar manner and used in the same way as described above for workflow change recommendations. Bradykinesia (i.e., slowness of movement) and rigidity (i.e., stiffness and resistance to movement) are common symptoms in neurodegenerative diseases such as Parkinson's disease and can also affect the performance of cognitive tests in the absence of cognitive impairment. Similar to the above, the detection of other types of movement disorders can be used to generate workflow change recommendations during evaluation such that (1) the test in which the movement disorder was detected is stopped if the movement disorder exceeds a certain threshold, (2) subsequent tests including the same movement element (i.e., drawing, writing, or speaking) are skipped from evaluation if the movement disorder exceeds a certain threshold, and / or (3) after evaluation such that the test in which the movement disorder was detected is removed from the overall cognitive model as well as individual domain test scores if the movement disorder exceeds a certain threshold.

[0136] In another embodiment, disease-specific reference scores can be used to correct cognitive test scores according to the determined tremor level. In that way, it is still possible to longitudinally evaluate cognitive function in a group of patients with tremors without completely removing the composite score test score calculation for the test in which tremors were detected.

[0137] The above-described embodiments of the present invention use one or more processors. The one or more processors can be arranged within a single housing device, structure, or unit, or may be distributed among multiple different devices, structures, or units. Thus, a reference to a processing configuration adapted or configured to perform a particular step or task can correspond to that step or task being performed by any one or more of a plurality of processing components, either alone or in combination. One skilled in the art will understand how such distributed processing devices can be implemented. The one or more processors can be implemented in various ways using software and / or hardware to perform the various required functions. The processor typically uses one or more microprocessors programmed using software (e.g., microcode) to perform the required functions. The processor may be implemented as a combination of dedicated hardware for performing some functions and one or more programmed microprocessors and associated circuitry for performing other functions.

[0138] Examples of circuits used in various embodiments of the present disclosure include, but are not limited to, conventional microprocessors, application-specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs).

[0139] In various embodiments, the processor may be associated with one or more storage media that are volatile and non-volatile computer memories such as, for example, RAM, PROM, EPROM, and EEPROM. This storage media may be encoded with one or more programs that perform the required functions when executed on one or more processors and / or controllers. The various storage media may be attached within the processor or controller, or may be transportable such that one or more programs stored on the storage media are loaded into the processor.

[0140] Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality.

[0141] A single processor or other unit can perform the functions of several items recited in the claims. A computer program can be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium, supplied together with or as part of other hardware, but can also be distributed in other forms via the Internet or other wired or wireless telecommunications systems.

[0142] The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used advantageously. When the term "adapted to" is used in a claim or in the specification, it means the same as the term "configured to".

[0143] Any reference signs in the claims should not be construed as limiting the scope.

Claims

Claim 1 A step of controlling a test device to perform a neuropsychological evaluation including motor elements on a subject, wherein the neuropsychological evaluation includes performing a plurality of cognitive tests according to an ordered schedule defined by a neuropsychological evaluation workflow, and processing individual test results according to processing operations defined by the neuropsychological evaluation workflow, and A step of automatically detecting one or more movement disorders during the execution of the neuropsychological evaluation; and A step of generating neuropsychological evaluation workflow change recommendations in response to the detection; and A step of outputting an output indicating the workflow change recommendation; and A method implemented on a computer, comprising: Claim 2 The method according to claim 1, further comprising a step of implementing the workflow change recommendation by changing the workflow according to the recommendation. Claim 3 The processing operation includes combining scores obtained for a plurality of individual cognitive tests of the neuropsychological evaluation into one or more cognitive domain scores according to a cognitive domain category pre-associated with each cognitive test, and optionally, the cognitive domain score includes one or more scores of memory, language processing, executive function, and visual-spatial processing. The workflow change recommendation includes a recommendation for modifying one or more cognitive domain scores. The method according to claim 1 or 2. Claim 4 In response to the detection of a movement disorder during the performance of a specific cognitive test, a user-perceivable output indicating the workflow change recommendation is generated during the neuropsychological evaluation, and the workflow change recommendation includes stopping the current cognitive test in which the movement disorder is detected. The method according to any one of claims 1 to 3. Claim 5 In response to the detection of a movement disorder during the neuropsychological evaluation, a user-perceivable output indicating the recommendation is generated during the evaluation, and the workflow change recommendation includes skipping subsequent tests scheduled according to the workflow involving the same motor elements as the cognitive test in which the movement disorder is detected. The method according to any one of claims 1 to 4. Claim 6 In response to the detection of a movement disorder during the neuropsychological evaluation, an output indicating the recommendation is generated after all the tests scheduled according to the workflow have been performed, and the workflow change recommendation includes that the score combination part of the processing operation is adjusted to compensate for the scores obtained for the individual cognitive tests in which the movement disorder was detected, and optionally, the scores for the tests in which a movement disorder was detected are excluded from the score combination calculation. The method according to any one of claims 1 to 5.

7. The method according to any one of claims 1 to 6, further comprising the step of determining a quantification of the severity of the detected movement disorder.

8. The implementation of at least one of the cognitive tests includes obtaining a test data set including one or more data variables associated with the performance of the test. The detection of one or more of the movement disorders includes processing at least a subset of the test data set by spectral analysis to derive a spectral power spectrum, and at least one detection of a movement disorder is based on the analysis of the spectral power spectrum. The method according to any one of claims 1 to 7.

9. The movement disorder to be detected includes motor tremors, and one or more of the data variables include data variables indicating the movement of a part of the subject's body as a function of time. The method according to claim 8.

10. The movement disorder to be detected includes vocal tremors, and one or more of the data variables include data variables indicating the audio acoustic data of the subject's speech input as a function of time. The method according to claim 8 or 9.

11. The severity of the tremor is quantified based on the spectral power spectrum from which it was derived. The method according to any one of claims 8 to 10.

12. The implementation of at least one of the cognitive tests includes obtaining a test data set including a plurality of data variables associated with the performance of the test. The detection of one or more of the movement disorders includes applying a Bayesian model to the plurality of data variables collected for the test, the Bayesian model being configured to output a probability score indicating the likelihood of the presence of a predefined movement disorder. The method according to any one of claims 1 to 11.

13. A plurality of Bayesian models are applied to generate respective probability scores for a plurality of different movement disorders, and the plurality of probability scores are combined into a combined probability score, and the workflow change recommendation is generated only in response to the combined probability score exceeding a predetermined threshold value. The method according to claim 12.

14. A computer program which is executed by a processor to cause the processor to execute the method according to any one of claims 1 to 13.

15. Having an input / output unit and one or more processors, wherein the processor controls a test device to perform a neuropsychological evaluation including movement elements on a subject via a control signal communicated through the input / output unit, and the neuropsychological evaluation includes performing a plurality of cognitive tests according to an ordered schedule defined by a neuropsychological evaluation workflow, and processing individual test results according to processing operations defined by the neuropsychological evaluation workflow; steps, automatically detecting one or more movement disorders during the performance of the neuropsychological evaluation; generating a neuropsychological evaluation workflow change recommendation in response to the detection; generating an output indicating the workflow change recommendation; A system configured to perform.

Images