From fall-risk assessment to fall detection: inertial sensors in the clinical routine and daily life

Falls are caused by complex interaction between multiple risk factors which may be modified by age, disease and environment. A variety of methods and tools for fall risk assessment have been proposed, but none of which is universally accepted. Existing tools are generally not capable of providing a quantitative predictive assessment of fall risk. The need for objective, cost-effective and clinically applicable methods would enable quantitative assessment of fall risk on a subject-specific basis. Tracking objectively falls risk could provide timely feedback about the effectiveness of administered interventions enabling intervention strategies to be modified or changed if found to be ineffective. Moreover, some of the fundamental factors leading to falls and what actually happens during a fall remain unclear. Objectively documented and measured falls are needed to improve knowledge of fall in order to develop more effective prevention strategies and prolong independent living. In the last decade, several research groups have developed sensor-based automatic or semi-automatic fall risk assessment tools using wearable inertial sensors. This approach may also serve to detect falls. At the moment, i) several fall-risk assessment studies based on inertial sensors, even if promising, lack of a biomechanical model-based approach which could provide accurate and more detailed measurements of interests (e.g., joint moments, forces) and ii) the number of published real-world fall data of older people in a real-world environment is minimal since most authors have used simulations with healthy volunteers as a surrogate for real-world falls. With these limitations in mind, this thesis aims i) to suggest a novel method for the kinematics and dynamics evaluation of functional motor tasks, often used in clinics for the fall-risk evaluation, through a body sensor network and a biomechanical approach and ii) to define the guidelines for a fall detection algorithm based on a real-world fall database availability.

Automatic induction of lexical features

This thesis concerns artificially intelligent natural language processing systems that are capable of learning the properties of lexical items (properties like verbal valency or inflectional class membership) autonomously while they are fulfilling their tasks for which they have been deployed in the first place. Many of these tasks require a deep analysis of language input, which can be characterized as a mapping of utterances in a given input C to a set S of linguistically motivated structures with the help of linguistic information encoded in a grammar G and a lexicon L: G + L + C → S (1) The idea that underlies intelligent lexical acquisition systems is to modify this schematic formula in such a way that the system is able to exploit the information encoded in S to create a new, improved version of the lexicon: G + L + S → L' (2) Moreover, the thesis claims that a system can only be considered intelligent if it does not just make maximum usage of the learning opportunities in C, but if it is also able to revise falsely acquired lexical knowledge. So, one of the central elements in this work is the formulation of a couple of criteria for intelligent lexical acquisition systems subsumed under one paradigm: the Learn-Alpha design rule. The thesis describes the design and quality of a prototype for such a system, whose acquisition components have been developed from scratch and built on top of one of the state-of-the-art Head-driven Phrase Structure Grammar (HPSG) processing systems. The quality of this prototype is investigated in a series of experiments, in which the system is fed with extracts of a large English corpus. While the idea of using machine-readable language input to automatically acquire lexical knowledge is not new, we are not aware of a system that fulfills Learn-Alpha and is able to deal with large corpora. To instance four major challenges of constructing such a system, it should be mentioned that a) the high number of possible structural descriptions caused by highly underspeci ed lexical entries demands for a parser with a very effective ambiguity management system, b) the automatic construction of concise lexical entries out of a bulk of observed lexical facts requires a special technique of data alignment, c) the reliability of these entries depends on the system's decision on whether it has seen 'enough' input and d) general properties of language might render some lexical features indeterminable if the system tries to acquire them with a too high precision. The cornerstone of this dissertation is the motivation and development of a general theory of automatic lexical acquisition that is applicable to every language and independent of any particular theory of grammar or lexicon. This work is divided into five chapters. The introductory chapter first contrasts three different and mutually incompatible approaches to (artificial) lexical acquisition: cue-based queries, head-lexicalized probabilistic context free grammars and learning by unification. Then the postulation of the Learn-Alpha design rule is presented. The second chapter outlines the theory that underlies Learn-Alpha and exposes all the related notions and concepts required for a proper understanding of artificial lexical acquisition. Chapter 3 develops the prototyped acquisition method, called ANALYZE-LEARN-REDUCE, a framework which implements Learn-Alpha. The fourth chapter presents the design and results of a bootstrapping experiment conducted on this prototype: lexeme detection, learning of verbal valency, categorization into nominal count/mass classes, selection of prepositions and sentential complements, among others. The thesis concludes with a review of the conclusions and motivation for further improvements as well as proposals for future research on the automatic induction of lexical features.

Non Invasive Tools for Early Detection of Autism Spectrum Disorders

Autism Spectrum Disorders (ASDs) describe a set of neurodevelopmental disorders. ASD represents a significant public health problem. Currently, ASDs are not diagnosed before the 2nd year of life but an early identification of ASDs would be crucial as interventions are much more effective than specific therapies starting in later childhood. To this aim, cheap an contact-less automatic approaches recently aroused great clinical interest. Among them, the cry and the movements of the newborn, both involving the central nervous system, are proposed as possible indicators of neurological disorders. This PhD work is a first step towards solving this challenging problem. An integrated system is presented enabling the recording of audio (crying) and video (movements) data of the newborn, their automatic analysis with innovative techniques for the extraction of clinically relevant parameters and their classification with data mining techniques. New robust algorithms were developed for the selection of the voiced parts of the cry signal, the estimation of acoustic parameters based on the wavelet transform and the analysis of the infant’s general movements (GMs) through a new body model for segmentation and 2D reconstruction. In addition to a thorough literature review this thesis presents the state of the art on these topics that shows that no studies exist concerning normative ranges for newborn infant cry in the first 6 months of life nor the correlation between cry and movements. Through the new automatic methods a population of control infants (“low-risk”, LR) was compared to a group of “high-risk” (HR) infants, i.e. siblings of children already diagnosed with ASD. A subset of LR infants clinically diagnosed as newborns with Typical Development (TD) and one affected by ASD were compared. The results show that the selected acoustic parameters allow good differentiation between the two groups. This result provides new perspectives both diagnostic and therapeutic.

Automatic Reduction of Artifacts in EEG-Signals

Electroencephalograms (EEG) are often contaminated with high amplitude artifacts limiting the usability of data. Methods that reduce these artifacts are often restricted to certain types of artifacts, require manual interaction or large training data sets. Within this paper we introduce a novel method, which is able to eliminate many different types of artifacts without manual intervention. The algorithm first decomposes the signal into different sub-band signals in order to isolate different types of artifacts into specific frequency bands. After signal decomposition with principal component analysis (PCA) an adaptive threshold is applied to eliminate components with high variance corresponding to the dominant artifact activity. Our results show that the algorithm is able to significantly reduce artifacts while preserving the EEG activity. Parameters for the algorithm do not have to be identified for every patient individually making the method a good candidate for preprocessing in automatic seizure detection and prediction algorithms.

Detecting offsets in GPS time series: first results from the Detection of Offsets in GPS Experiment

The accuracy of Global Positioning System (GPS) time series is degraded by the presence of offsets. To assess the effectiveness of methods that detect and remove these offsets, we designed and managed the Detection of Offsets in GPS Experiment. We simulated time series that mimicked realistic GPS data consisting of a velocity component, offsets, white and flicker noises (1/f spectrum noises) composed in an additive model. The data set was made available to the GPS analysis community without revealing the offsets, and several groups conducted blind tests with a range of detection approaches. The results show that, at present, manual methods (where offsets are hand picked) almost always give better results than automated or semi‒automated methods (two automated methods give quite similar velocity bias as the best manual solutions). For instance, the fifth percentile range (5% to 95%) in velocity bias for automated approaches is equal to 4.2 mm/year (most commonly ±0.4 mm/yr from the truth), whereas it is equal to 1.8 mm/yr for the manual solutions (most commonly 0.2 mm/yr from the truth). The magnitude of offsets detectable by manual solutions is smaller than for automated solutions, with the smallest detectable offset for the best manual and automatic solutions equal to 5 mm and 8 mm, respectively. Assuming the simulated time series noise levels are representative of real GPS time series, robust geophysical interpretation of individual site velocities lower than 0.2–0.4 mm/yr is therefore certainly not robust, although a limit of nearer 1 mm/yr would be a more conservative choice. Further work to improve offset detection in GPS coordinates time series is required before we can routinely interpret sub‒mm/yr velocities for single GPS stations.

Fully automatic X-ray image segmentation via joint estimation of image displacements.

We propose a new method for fully-automatic landmark detection and shape segmentation in X-ray images. Our algorithm works by estimating the displacements from image patches to the (unknown) landmark positions and then integrating them via voting. The fundamental contribution is that, we jointly estimate the displacements from all patches to multiple landmarks together, by considering not only the training data but also geometric constraints on the test image. The various constraints constitute a convex objective function that can be solved efficiently. Validated on three challenging datasets, our method achieves high accuracy in landmark detection, and, combined with statistical shape model, gives a better performance in shape segmentation compared to the state-of-the-art methods.

Software-based detection of atrial fibrillation in long-term ECGs

Background Atrial fibrillation (AF) is common and may have severe consequences. Continuous long-term electrocardiogram (ECG) is widely used for AF screening. Recently, commercial ECG analysis software was launched, which automatically detects AF in long-term ECGs. It has been claimed that such tools offer reliable AF screening and save time for ECG analysis. However, this has not been investigated in a real-life patient cohort. Objective To investigate the performance of automatic software-based screening for AF in long-term ECGs. Methods Two independent physicians manually screened 22,601 hours of continuous long-term ECGs from 150 patients for AF. Presence, number, and duration of AF episodes were registered. Subsequently, the recordings were screened for AF by an established ECG analysis software (Pathfinder SL), and its performance was validated against the thorough manual analysis (gold standard). Results Sensitivity and specificity for AF detection was 98.5% (95% confidence interval 91.72%–99.96%) and 80.21% (95% confidence interval 70.83%–87.64%), respectively. Software-based AF detection was inferior to manual analysis by physicians (P < .0001). Median AF duration was underestimated (19.4 hours vs 22.1 hours; P < .001) and median number of AF episodes was overestimated (32 episodes vs 2 episodes; P < .001) by the software. In comparison to extensive quantitative manual ECG analysis, software-based analysis saved time (2 minutes vs 19 minutes; P < .001). Conclusion Owing to its high sensitivity and ability to save time, software-based ECG analysis may be used as a screening tool for AF. An additional manual confirmatory analysis may be required to reduce the number of false-positive findings.

Comparison of gadoteric acid and gadobutrol for detection as well as morphologic and dynamic characterization of lesions on breast dynamic contrast-enhanced magnetic resonance imaging

OBJECTIVE In contrast to conventional breast imaging techniques, one major diagnostic benefit of breast magnetic resonance imaging (MRI) is the simultaneous acquisition of morphologic and dynamic enhancement characteristics, which are based on angiogenesis and therefore provide insights into tumor pathophysiology. The aim of this investigation was to intraindividually compare 2 macrocyclic MRI contrast agents, with low risk for nephrogenic systemic fibrosis, in the morphologic and dynamic characterization of histologically verified mass breast lesions, analyzed by blinded human evaluation and a fully automatic computer-assisted diagnosis (CAD) technique. MATERIALS AND METHODS Institutional review board approval and patient informed consent were obtained. In this prospective, single-center study, 45 women with 51 histopathologically verified (41 malignant, 10 benign) mass lesions underwent 2 identical examinations at 1.5 T (mean time interval, 2.1 days) with 0.1-mmol kg doses of gadoteric acid and gadobutrol. All magnetic resonance images were visually evaluated by 2 experienced, blinded breast radiologists in consensus and by an automatic CAD system, whereas the morphologic and dynamic characterization as well as the final human classification of lesions were performed based on the categories of the Breast imaging reporting and data system MRI atlas. Lesions were also classified by defining their probability of malignancy (morpho-dynamic index; 0%-100%) by the CAD system. Imaging results were correlated with histopathology as gold standard. RESULTS The CAD system coded 49 of 51 lesions with gadoteric acid and gadobutrol (detection rate, 96.1%); initial signal increase was significantly higher for gadobutrol than for gadoteric acid for all and the malignant coded lesions (P < 0.05). Gadoteric acid resulted in more postinitial washout curves and fewer continuous increases of all and the malignant lesions compared with gadobutrol (CAD hot spot regions, P < 0.05). Morphologically, the margins of the malignancies were different between the 2 agents, whereas gadobutrol demonstrated more spiculated and fewer smooth margins (P < 0.05). Lesion classifications by the human observers and by the morpho-dynamic index compared with the histopathologic results did not significantly differ between gadoteric acid and gadobutrol. CONCLUSIONS Macrocyclic contrast media can be reliably used for breast dynamic contrast-enhanced MRI. However, gadoteric acid and gadobutrol differed in some dynamic and morphologic characterization of histologically verified breast lesions in an intraindividual, comparison. Besides the standardization of technical parameters and imaging evaluation of breast MRI, the standardization of the applied contrast medium seems to be important to receive best comparable MRI interpretation.

Fully Automatic CT Segmentation for Computer-Assisted Pre-operative Planning of Hip Arthroscopy

Extraction of both pelvic and femoral surface models of a hip joint from CT data for computer-assisted pre-operative planning of hip arthroscopy is addressed. We present a method for a fully automatic image segmentation of a hip joint. Our method works by combining fast random forest (RF) regression based landmark detection, atlas-based segmentation, with articulated statistical shape model (aSSM) based hip joint reconstruction. The two fundamental contributions of our method are: (1) An improved fast Gaussian transform (IFGT) is used within the RF regression framework for a fast and accurate landmark detection, which then allows for a fully automatic initialization of the atlas-based segmentation; and (2) aSSM based fitting is used to preserve hip joint structure and to avoid penetration between the pelvic and femoral models. Validation on 30 hip CT images show that our method achieves high performance in segmenting pelvis, left proximal femur, and right proximal femur surfaces with an average accuracy of 0.59 mm, 0.62 mm, and 0.58 mm, respectively.

Fully automatic segmentation of AP pelvis X-rays via random forest regression with efficient feature selection and hierarchical sparse shape composition

In clinical practice, traditional X-ray radiography is widely used, and knowledge of landmarks and contours in anteroposterior (AP) pelvis X-rays is invaluable for computer aided diagnosis, hip surgery planning and image-guided interventions. This paper presents a fully automatic approach for landmark detection and shape segmentation of both pelvis and femur in conventional AP X-ray images. Our approach is based on the framework of landmark detection via Random Forest (RF) regression and shape regularization via hierarchical sparse shape composition. We propose a visual feature FL-HoG (Flexible- Level Histogram of Oriented Gradients) and a feature selection algorithm based on trace radio optimization to improve the robustness and the efficacy of RF-based landmark detection. The landmark detection result is then used in a hierarchical sparse shape composition framework for shape regularization. Finally, the extracted shape contour is fine-tuned by a post-processing step based on low level image features. The experimental results demonstrate that our feature selection algorithm reduces the feature dimension in a factor of 40 and improves both training and test efficiency. Further experiments conducted on 436 clinical AP pelvis X-rays show that our approach achieves an average point-to-curve error around 1.2 mm for femur and 1.9 mm for pelvis.

FACTS: Fully Automatic CT Segmentation of a Hip Joint

Extraction of surface models of a hip joint from CT data is a pre-requisite step for computer assisted diagnosis and planning (CADP) of periacetabular osteotomy (PAO). Most of existing CADP systems are based on manual segmentation, which is time-consuming and hard to achieve reproducible results. In this paper, we present a Fully Automatic CT Segmentation (FACTS) approach to simultaneously extract both pelvic and femoral models. Our approach works by combining fast random forest (RF) regression based landmark detection, multi-atlas based segmentation, with articulated statistical shape model (aSSM) based fitting. The two fundamental contributions of our approach are: (1) an improved fast Gaussian transform (IFGT) is used within the RF regression framework for a fast and accurate landmark detection, which then allows for a fully automatic initialization of the multi-atlas based segmentation; and (2) aSSM based fitting is used to preserve hip joint structure and to avoid penetration between the pelvic and femoral models. Taking manual segmentation as the ground truth, we evaluated the present approach on 30 hip CT images (60 hips) with a 6-fold cross validation. When the present approach was compared to manual segmentation, a mean segmentation accuracy of 0.40, 0.36, and 0.36 mm was found for the pelvis, the left proximal femur, and the right proximal femur, respectively. When the models derived from both segmentations were used to compute the PAO diagnosis parameters, a difference of 2.0 ± 1.5°, 2.1 ± 1.6°, and 3.5 ± 2.3% were found for anteversion, inclination, and acetabular coverage, respectively. The achieved accuracy is regarded as clinically accurate enough for our target applications.

Evaluation and Comparison of Anatomical Landmark Detection Methods for Cephalometric X-Ray Images: A Grand Challenge

Cephalometric analysis is an essential clinical and research tool in orthodontics for the orthodontic analysis and treatment planning. This paper presents the evaluation of the methods submitted to the Automatic Cephalometric X-Ray Landmark Detection Challenge, held at the IEEE International Symposium on Biomedical Imaging 2014 with an on-site competition. The challenge was set to explore and compare automatic landmark detection methods in application to cephalometric X-ray images. Methods were evaluated on a common database including cephalograms of 300 patients aged six to 60 years, collected from the Dental Department, Tri-Service General Hospital, Taiwan, and manually marked anatomical landmarks as the ground truth data, generated by two experienced medical doctors. Quantitative evaluation was performed to compare the results of a representative selection of current methods submitted to the challenge. Experimental results show that three methods are able to achieve detection rates greater than 80% using the 4 mm precision range, but only one method achieves a detection rate greater than 70% using the 2 mm precision range, which is the acceptable precision range in clinical practice. The study provides insights into the performance of different landmark detection approaches under real-world conditions and highlights achievements and limitations of current image analysis techniques.

A New Approach for Automatic Removal of Movement Artifacts in Near-Infrared Spectroscopy Time Series by Means of Acceleration Data

Abstract: Near-infrared spectroscopy (NIRS) enables the non-invasive measurement of changes in hemodynamics and oxygenation in tissue. Changes in light-coupling due to movement of the subject can cause movement artifacts (MAs) in the recorded signals. Several methods have been developed so far that facilitate the detection and reduction of MAs in the data. However, due to fixed parameter values (e.g., global threshold) none of these methods are perfectly suitable for long-term (i.e., hours) recordings or were not time-effective when applied to large datasets. We aimed to overcome these limitations by automation, i.e., data adaptive thresholding specifically designed for long-term measurements, and by introducing a stable long-term signal reconstruction. Our new technique (“acceleration-based movement artifact reduction algorithm”, AMARA) is based on combining two methods: the “movement artifact reduction algorithm” (MARA, Scholkmann et al. Phys. Meas. 2010, 31, 649–662), and the “accelerometer-based motion artifact removal” (ABAMAR, Virtanen et al. J. Biomed. Opt. 2011, 16, 087005). We describe AMARA in detail and report about successful validation of the algorithm using empirical NIRS data, measured over the prefrontal cortex in adolescents during sleep. In addition, we compared the performance of AMARA to that of MARA and ABAMAR based on validation data.

Fully Automatic Segmentation of Hip CT Images

Automatic segmentation of the hip joint with pelvis and proximal femur surfaces from CT images is essential for orthopedic diagnosis and surgery. It remains challenging due to the narrowness of hip joint space, where the adjacent surfaces of acetabulum and femoral head are hardly distinguished from each other. This chapter presents a fully automatic method to segment pelvic and proximal femoral surfaces from hip CT images. A coarse-to-fine strategy was proposed to combine multi-atlas segmentation with graph-based surface detection. The multi-atlas segmentation step seeks to coarsely extract the entire hip joint region. It uses automatically detected anatomical landmarks to initialize and select the atlas and accelerate the segmentation. The graph based surface detection is to refine the coarsely segmented hip joint region. It aims at completely and efficiently separate the adjacent surfaces of the acetabulum and the femoral head while preserving the hip joint structure. The proposed strategy was evaluated on 30 hip CT images and provided an average accuracy of 0.55, 0.54, and 0.50 mm for segmenting the pelvis, the left and right proximal femurs, respectively.

Dish Detection and Segmentation for Dietary Assessment on Smartphones

Diet-related chronic diseases severely affect personal and global health. However, managing or treating these diseases currently requires long training and high personal involvement to succeed. Computer vision systems could assist with the assessment of diet by detecting and recognizing different foods and their portions in images. We propose novel methods for detecting a dish in an image and segmenting its contents with and without user interaction. All methods were evaluated on a database of over 1600 manually annotated images. The dish detection scored an average of 99% accuracy with a .2s/image run time, while the automatic and semi-automatic dish segmentation methods reached average accuracies of 88% and 91% respectively, with an average run time of .5s/image, outperforming competing solutions.