Skip to main content

Hilkka Liedes: Data-driven methods can predict progression of Alzheimer’s disease

Tampere University
LocationTietotalo auditorium TB109, Hervanta Campus, Korkeakoulunkatu 7, and remote connection
17.2.2023 10.00–14.00
LanguageEnglish
Entrance feeFree of charge
Hilkka Liedes
Data-driven analysis and visualization methods may help with interpretation and exploitation of large amounts of heterogeneous data. In her doctoral dissertation, MHSc Hilkka Liedes developed and validated data-driven methods for prediction and monitoring of progression of Alzheimer’s disease in different phases of the disease spectrum.

Due to ageing of the population, the number of people with Alzheimer’s disease increases considerably. Even though the disease cannot be cured yet, certain medical and lifestyle interventions can delay its progression and alleviate symptoms, if they are started already at an early phase of the disease.

However, early diagnosis and prognosis is challenging because pathological changes in the brain start already years before the symptoms appear. In addition, there is no simple and single test for Alzheimer’s disease. Instead, many different examinations are performed, and clinicians need to combine and interpret their results.

Data-driven analysis and visualization methods may help with interpretation and utilization of large amounts of heterogeneous patient data.

In her doctoral dissertation, Hilkka Liedes developed and validated data-driven methods for prediction and monitoring of progression of Alzheimer’s disease in different phases of the disease spectrum.

She utilized data from existing databases collected in three European countries, North America and Australia. The databases contained background information, cognitive and neuropsychological test results, magnetic resonance imaging scans, and results from cerebrospinal fluid samples and from genetic testing.

Combination of several data modalities provided more accurate results

The results showed that the data-driven methods can be used for predicting and monitoring progression of Alzheimer’s disease from the mildest stages to more advanced stages. Combining information from several data modalities provided better prediction performance than using individual data modalities alone.

It was also observed that it is extremely important to validate the developed methods with independent validation cohorts which have not been used in the development phase. This will provide more reliable results and insights to limitations of the methods.

— In different countries, different patient examination methods and diagnostic criteria are used. Thus, introduction of these methods to different environments and countries may require harmonization of patient examination methods and diagnostic criteria, Liedes points out.

The developed methods can support a clinician in assessment of their patient’s state and in early diagnosis of Alzheimer’s disease, which enables timely treatment and thus positively influences the patient’s quality of life.

— This will in turn bring cost savings to society as patients can live longer independently at home without need for intensive nursing, Liedes added.

Furthermore, the methods can potentially aid in identifying suitable patients for Alzheimer’s disease drug trials.

Hilkka Liedes lives currently in Oulu and continues her work at VTT Technical Research Centre of Finland Ltd in the field of health data analytics.

Public defence on Friday 17 February

The doctoral dissertation of MHSc Hilkka Liedes in the field of medical engineering titled Prediction and Monitoring of Progression of Alzheimer’s Disease: Multivariable approaches for decision support will be publicly examined at the Faculty of Medicine and Health Technology of Tampere University on Friday 17 February 2023 at 12 o’clock at Tietotalo auditorium TB109 (Hervanta Campus, Korkeakoulunkatu 7, Tampere). The Opponent will be Professor Natasha Maurits, University of Groningen, The Netherlands. The Custos will be Professor Mark van Gils, Faculty of Medicine and Health Technology, Tampere University.  

The doctoral dissertation is available online.

The public defence can be followed via remote connection.


Photograph: Petri Liedes