BPI cluster meeting

On 7th of May, 12:30-13:30 in Pav K 16 Guillaume Zamora will present his work on “A clinical decision support system by using wrist-worn smartphone tremor measurements”.

Everyone is welcome!

 

Abstract:

Background: Tremor related diseases affect millions of people around the world, hindering various everyday life tasks, such as holding a glass of water. Tremor severity assessment is an important element for the diagnosis and treatment decision making process. For decades, subjective clinical rating scales were mostly performed. Recently, remarkable attention around computerized tremor analysis has grown. While dedicated devices are expensive and not practical for the everyday use, smartphone applications are promising. Previous studies on Parkinson’s disease or Essential Tremor mostly classified the Fahn score rating scale. Objective: Using machine learning techniques to
regress tremor severity observed by clinicians (ETRS) and patients (QUEST), and give the research accessibility and new insights that would later lead to decision making process improvements.
Methods: Five wrist-worn different tests were performed on 20 Essential Tremor patients from the open-source TREMOR12 iPhone/iWatch compatible application. Linear displacements and joint rotations are measured from in-device accelerometer and gyroscope. From these signals, time, frequency and time-frequency domain tools are used to extract the following features: dominant frequency, dominant magnitude, signal RMS, signal period and the power growth during the test.
Results: While the study demonstrates good predictive power, its feature extraction shows to bring improvements when compared to previous close setting studies. Conclusion: This study gives the research new directions and tools in order to perform further investigations around tremor severity evaluation. Smartphone sensors improvement in the following years, research on the best predicted variable to use and larger data collection may lead to very robust models, measuring rapidly, accurately and more objectively tremor severity than clinical rating scales.
Key words: movement disorder, tremor severity, smartphone, feature extraction, regression

BPI cluster meeting – presentation by Yingqian Zhang

On Wednesday, 17.05.2017, 12:30-13:30 in Pav K16
Yingqian Zhang will present her talk on
“Learning decision trees with flexible constraints and objectives using integer optimization”

Abstract:

We encode the problem of learning the optimal decision tree of a given depth as an integer optimization problem. We show experimentally that our method (DTIP) can be used to learn good trees up to depth 5 from data sets of size up to 1000. In addition to being efficient, our new formulation allows for a lot of flexibility. Experiments show that we can use the trees learned from any existing decision tree algorithms as starting solutions and improve the trees using DTIP. Moreover, the proposed formulation allows us to easily create decision trees with different optimization objectives instead of accuracy and error, and constraints can be added explicitly during the tree construction phase. We show how this flexibility can be used to learn discrimination-aware classification trees, to improve learning from imbalanced data, and to learn trees that minimise false positive/negative errors.

Everyone is welcome!