Dalarna University's logo and link to the university's website

Background Dementia disorders are affecting millions of people globally, characterized by memory loss, communication difficulties, and motor function decline. Accurate and early dementia detection is crucial for effective management and treatment. Gait analysis offers a non-invasive method for dementia detection by identifying subtle changes in walking patterns that often precede cognitive symptoms.

Objective This study aims to evaluate the clinical utility of video-based gait analysis using the Timed Up and Go (TUG) test under single and dual-task conditions (TUGdt) for distinguishing individuals with dementia disorders from healthy controls (HCs).

Method The study implemented three machine learning models: Support Vector Machine (SVM), Logistic Regression (LR), and Random Forest (RF), to discriminate between persons with dementia and HCs. The dataset consists of a cohort of 64 people with dementia (47 with Alzheimer's disease) and 67 HCs. The participants performed the TUG test as a single and dual-task (TUGdt). In the TUGdt, participants performed the TUG test while simultaneously completing an additional cognitive task (i.e., animal naming (TUGdt-NA) or reciting months in reverse order (TUGdt-MB)).

Results The results showed that dual-task classification outperformed the single-task. The SVM algorithm achieved the highest accuracy in the TUGdt-NA task (accuracy of 87% ± 5.1 and recall of 86.6% ± 3.2) using 5-fold cross-validation and accuracy of 85.5% and recall of 89.5% using Leave-One-Out Cross-Validation (LOOCV) in the TUGdt-MB task.

Conclusions In summary, video-based gait features effectively distinguish people with dementia from HCs, particularly under dual-tasking, offering cost-effective, automated, and non-invasive pre-screening to complement clinical assessments.

Forskningsanknytning är en central men svårfångad aspekt av utbildningskvalitet, särskilt i tekniska utbildningar. Vid Högskolan Dalarna har en kollegial metod för programanalys utvecklats för att synliggöra och stärka forskningsanknytningen och samtidigt fungera som en plattform för kollegialt lärande. Metoden kombinerar ett programanalysverktyg med workshopbaserad reflektion utifrån en fyrfältsmodell för forskningsförankring. Resultaten visar att arbetet har skapat samsyn och ökat engagemang bland lärare samt tydliggjort behovet av att utveckla kunskap om hur forskningsanknytning kan omsättas i praktiken. Metoden bidrar därmed till kvalitetsutveckling genom gemensam förståelse och reflektion i lärarlag. 

Dementia is a neurodegenerative disorder that affects millions of individuals around the world. Early diagnosis of dementia is crucial to provide timely intervention and management. Gait analysis offers insights into the cognitive impairments, which are early signs of dementia. This study aims to extract several relevant features for gait analysis and utilize these features in machine learning algorithms, specifically Extreme Gradient Boosting (XGBoost), Random Forest (RF), and Support Vector Machine (SVM), for multiclass classification of dementia (Mild cognitive impairment (MCI), subjective cognitive impairment (SCI), dementia, and healthy controls (HCs)). The dataset used in this study consists of videos of 64 people with dementia, 63 with MCI, 64 with SCI, and 67 HCs. Participants performed the Timed Up and Go (TUG) test under single-task and dualtask conditions (either animal naming (TUGdt-NA) or reciting months in reverse order (TUGdt-MB)). The findings reveal that dual-task gait consistently outperformed single task. Moreover, the XGBoost algorithm achieved the highest accuracy of 87 % in the naming animals dual-task (TUGdtNA).

BACKGROUND: Dementia is a general term for several progressive neurodegenerative disorders including Alzheimer's disease. Timely and accurate detection is crucial for early intervention. Advancements in artificial intelligence present significant potential for using machine learning to aid in early detection.

OBJECTIVE: Summarize the state-of-the-art machine learning-based approaches for dementia prediction, focusing on non-invasive methods, as the burden on the patients is lower. Specifically, the analysis of gait and speech performance can offer insights into cognitive health through clinically cost-effective screening methods.

METHODS: A systematic literature review was conducted following the PRISMA protocol (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). The search was performed on three electronic databases (Scopus, Web of Science, and PubMed) to identify the relevant studies published between 2017 to 2022. A total of 40 papers were selected for review.

RESULTS: The most common machine learning methods employed were support vector machine followed by deep learning. Studies suggested the use of multimodal approaches as they can provide comprehensive and better prediction performance. Deep learning application in gait studies is still in the early stages as few studies have applied it. Moreover, including features of whole body movement contribute to better classification accuracy. Regarding speech studies, the combination of different parameters (acoustic, linguistic, cognitive testing) produced better results.

CONCLUSIONS: The review highlights the potential of machine learning, particularly non-invasive approaches, in the early prediction of dementia. The comparable prediction accuracies of manual and automatic speech analysis indicate an imminent fully automated approach for dementia detection.

Acute traumatic brain injury (TBI) is associated with substantial abnormalities in lipid biology, including changes in the structural lipids that are present in the myelin in the brain. We investigated the relationship between traumatic microstructural changes in white matter from magnetic resonance imaging (MRI) and quantitative lipidomic changes from blood serum. The study cohort included 103 patients from the Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI) study. Diffusion tensor fitting generated fractional anisotropy (FA) and mean diffusivity (MD) maps for the MRI scans while ultra-high-performance liquid chromatography quadrupole time-of-flight mass spectrometry was applied to analyze the lipidome. Increasing severity of TBI was associated with higher MD and lower FA values, which scaled with different lipidomic signatures. There appears to be consistent patterns of lipid changes associating with the specific microstructure changes in the CNS white matter, but also regional specificity, suggesting that blood-based lipidomics may provide an insight into the underlying pathophysiology of TBI.

Weather Normalized Models (WNMs) are modeling methods used for assessing air contaminants under a business-as-usual (BAU) assumption. Therefore, WNMs are used to assess the impact of many events on urban pollution. Recently, different approaches have been implemented to develop WNMs and quantify the lockdown effects of COVID-19 on air quality, including Machine Learning (ML). However, more advanced methods, such as Deep Learning (DL), have never been applied for developing WNMs. In this study, we proposed WNMs based on DL algorithms, aiming to test five DL architectures and compare their performances to a recent ML approach, namely Gradient Boosting Machine (GBM). The concentrations of five air pollutants (CO, NO₂, PM_2.5, SO₂, and O₃) are studied in the city of Quito, Ecuador. The results show that Long-Short Term Memory (LSTM) and Bidirectional Recurrent Neural Network (BiRNN) outperform the other algorithms and, consequently, are recommended as appropriate WNMs to quantify the effects of the lockdowns on air pollution. Furthermore, examining the variable importance in the LSTM and BiRNN models, we identify that the most relevant temporal and meteorological features for predicting air quality are Hours (time of day), Index (1 is the first collected data and increases by one after each instance), Julian Day (day of the year), Relative Humidity, Wind Speed, and Solar Radiation. During the full lockdown, the concentration of most pollutants has decreased drastically: −48.75%, for CO, −45.76%, for SO₂, −42.17%, for PM_2.5, and −63.98%, for NO₂. The reduction of this latter gas has induced an increase of O₃ by +26.54%.

AIM: To construct a Treatment Response Index from Multiple Sensors (TRIMS) for quantification of motor state in patients with Parkinson's disease (PD) during a single levodopa dose. Another aim was to compare TRIMS to sensor indexes derived from individual motor tasks.

METHOD: Nineteen PD patients performed three motor tests including leg agility, pronation-supination movement of hands, and walking in a clinic while wearing inertial measurement unit sensors on their wrists and ankles. They performed the tests repeatedly before and after taking 150% of their individual oral levodopa-carbidopa equivalent morning dose.Three neurologists blinded to treatment status, viewed patients' videos and rated their motor symptoms, dyskinesia, overall motor state based on selected items of Unified PD Rating Scale (UPDRS) part III, Dyskinesia scale, and Treatment Response Scale (TRS). To build TRIMS, out of initially 178 extracted features from upper- and lower-limbs data, 39 features were selected by stepwise regression method and were used as input to support vector machines to be mapped to mean reference TRS scores using 10-fold cross-validation method. Test-retest reliability, responsiveness to medication, and correlation to TRS as well as other UPDRS items were evaluated for TRIMS.

RESULTS: The correlation of TRIMS with TRS was 0.93. TRIMS had good test-retest reliability (ICC = 0.83). Responsiveness of the TRIMS to medication was good compared to TRS indicating its power in capturing the treatment effects. TRIMS was highly correlated to dyskinesia (R = 0.85), bradykinesia (R = 0.84) and gait (R = 0.79) UPDRS items. Correlation of sensor index from the upper-limb to TRS was 0.89.

CONCLUSION: Using the fusion of upper- and lower-limbs sensor data to construct TRIMS provided accurate PD motor states estimation and responsive to treatment. In addition, quantification of upper-limb sensor data during walking test provided strong results.

Parkinson’s disease (PD) is the second most common neurodegenerative disease. Levodopa is mainly used to manage the motor symptoms of PD. However, disease progression and long-term use of levodopa cause reduced medication efficacy and side effects. When that happens, precise individualized dosing schedules are required.

This doctoral thesis in the field of Micro-data analysis introduces an end-to-end solution for the automation of the pharmacological management of PD with levodopa, and offers some insight on levodopa pharmacodynamics. For that purpose, an algorithm that derives objective ratings for the patients’ motor function through wearable sensors is introduced, a method to construct individual patient profiles is developed, and two dosing algorithms for oral and intestinal administration of levodopa are presented. Data from five different sources were used to develop the methods and evaluate the performance of the proposed algorithms.

The dose automation algorithms can work both with clinical and objective ratings (through wearable devices), and their application was evaluated against dosing adjustments from movement disorders experts. Both dosing algorithms showed promise and their dosing suggestions were similar to those of the clinicians.

The objective ratings algorithm had good test-retest reliability and its application during a clinical study was successful. Furthermore, the method of fitting individual patient models was robust and worked well with the objective ratings algorithm. Finally, a study was carried out that showed that about half the patients on levodopa treatment show reduced response during the afternoon hours, pointing to the need for more precise modelling of levodopa pharmacodynamics.

INTRODUCTION: A treatment response objective index (TRIS) was previously developed based on sensor data from pronation-supination tests. This study aimed to examine the performance of TRIS for medication effects in a new population sample with Parkinson's disease (PD) and its usefulness for constructing individual dose-response models.

METHODS: Twenty-five patients with PD performed a series of tasks throughout a levodopa challenge while wearing sensors. TRIS was used to determine motor changes in pronation-supination tests following a single levodopa dose, and was compared to clinical ratings including the Treatment Response Scale (TRS) and six sub-items of the UPDRS part III.

RESULTS: As expected, correlations between TRIS and clinical ratings were lower in the new population than in the initial study. TRIS was still significantly correlated to TRS (r_s = 0.23, P < 0.001) with a root mean square error (RMSE) of 1.33. For the patients (n = 17) with a good levodopa response and clear motor fluctuations, a stronger correlation was found (r_s = 0.38, RMSE = 1.29, P < 0.001). The mean TRIS increased significantly when patients went from the practically defined off to their best on state (P = 0.024). Individual dose-response models could be fitted for more participants when TRIS was used for modelling than when TRS ratings were used.

CONCLUSION: The objective sensor index shows promise for constructing individual dose-response models, but further evaluations and retraining of the TRIS algorithm are desirable to improve its performance and to ensure its clinical effectiveness.