All the presentations from the ‘Valais Workshop on Artificial Intelligence: Reproducibility in Research’ are now available in the following webcast.
The complete workshop agenda is available here.
The journal article ‘3D Solid Texture Classification Using Locally-Oriented Wavelet Transforms’ is now available under the ‘Early Access’ area in IEEEXplore.
The article will appear on the April issue of the IEEE Transactions on Image Processing journal.
Abstract:
The Chapter ‘Big data in medical imaging, forensics and beyond’ by Henning Müller has now been published in the book Forensigraphie – Möglichkeiten und Grenzen IT-gestützter klinisch-forensischer Bildgebung
Info
The paper ‘Making-sense of large data sets without annotations: analyzing age-related correlations from lung CT scans’ by Dicente et al. won the Robert F. Wagner Best student paper award for the track Imaging Informatics for Healthcare, Research and Applications.
The paper can be found in our publications page.
The paper ‘3-D Solid Texture Classification Using Locally-Oriented Wavelet Transforms’ by Dicente et al. has been accepted for publication at IEEE Transactions on Image Processing.
Abstract
Many image acquisition techniques used in biomedical imaging, material analysis, or structural geology are capable to acquire 3–D solid images. Computational analysis of these images is complex but necessary, since it is difficult for humans to visualize and quantify their detailed 3–D content. One of the most common methods to analyze 3–D data is to characterize the volumetric texture patterns. Texture analysis generally consists of encoding the local organization of image scales and directions, which can be extremely diverse in 3–D. Current state–of–the–art techniques face many challenges when working with 3–D solid texture, where most approaches are not able to consistently characterize both scale and directional information. 3–D Riesz–wavelets can deal with both properties. One key property of Riesz filterbanks is steerability, which can be used to locally align the filters and to compare textures with arbitrary (local) orientations. This paper compares three local alignment criteria for higher–order 3–D Riesz–wavelet transforms. The estimations of local texture orientations are based on higher–order extensions of regularized structure tensors. An experimental evaluation of the proposed methods for the classification of synthetic 3–D solid textures with alterations (e.g., rotations, noise) demonstrated the importance of local directional information for robust and accurate solid texture recognition. These alignment methods improved the accuracy of the unaligned Riesz descriptors by up to 0.63, from 0.32 to 0.95 over 1 in the rotated data, which are better than the other techniques tested on the same database.
The paper ‘Making Sense of Large Data Sets without Annotations:Analyzing Age-related Correlations from Lung CT Scans’ by Yashin Dicente et al. is one of the finalists for the Robert F. Wagner Best Student paper Award at SPIE2017. The conference will take place from the 11-16 February 2017 in Orlando, Florida, US.
Abstract:
The analysis of large data sets can help to gain knowledge about specic organs or on specic diseases, just as big data analysis does in many non-medical areas. This article aims to gain information from 3D volumes, so the visual content of lung CT scans of a large number of patients. In the case of the described data set, only little annotation is available on the patients that were all part of an ongoing screening program and besides age and gender no information on the patient and the ndings was available for this work. This is a scenario that can happen regularly as image data sets are produced and become available in increasingly large quantities but manual annotations are often not available and also clinical data such as text reports are often harder to share. We extracted a set of visual features from 12,414 CT scans of 9,348 patients that had CT scans of the lung taken in the context of a national lung screening program in Belarus. Lung elds were segmented by two segmentation algorithms and only cases where both algorithms were able to nd left and right lung and had a Dice coecient above 0.95 were analyzed. This assures that only segmentations of good quality were used to extract features of the lung. Patients ranged in age from 0 to 106 years. Data analysis shows that age can be predicted with a fairly high accuracy for persons under 15 years. Relatively good results were also obtained between 30 and 65
years where a steady trend is seen. For young adults and older people the results are not as good as variability is very high in these groups. Several visualizations of the data show the evolution patters of the lung texture, size and density with age. The experiments allow learning the evolution of the lung and the gained results show that even with limited meta{data we can extract interesting information from large{scale visual data. These age-related changes (for example of the lung volume, the density histogram of the tissue) can also be taken into account for the interpretation of new cases. The database used includes patients that had suspicions on a chest X-ray, so it is not a group of healthy people, and only tendencies and not a model of a healthy lung at a specic age can be derived.
Prof. Henning Müller is part of the Management Committe of the approved European Cooperation in Science and Technology (COST) action European network for translational research in children’s and adult interstitial lung disease
COST Actions are a flexible, fast, effective and efficient networking instrument for researchers, engineers and scholars to cooperate and coordinate nationally funded research activities. COST Actions allow European researchers to jointly develop their own ideas in any science and technology field.
Description
Interstitial lung disease in children (chILD) is a term that describes a collection of more than 200 rare lung disorders. It is a heterogeneous group of non-neoplastic disorders resulting from damage by varying patterns of inflammation and fibrosis with the interstitium as the primary site of injury. As with other orphan diseases, chILD data is lacking on the natural course, phenotypic variability, associations with genotype, and effectiveness of treatments. The disease course is very variable, and depending on more than just the underlying cause; for example, within a given family, the phenotypic variability of chILD such as surfactant protein C mutation is huge. The rarity of individual chILDs contributes to a lack of randomised control trial data on effectiveness of treatments. Management strategies derive from other diseases or are based on physicians experience and remain controversial.
This Action will create a pan-Europe-led network of multidisciplinary clinicians (adult and paediatric), scientists, and patients and their families with the aim of accurate and early diagnosis with structured, potentially personalised, management and therapies. The Action will stimulate and coordinate multidisciplinary research in chILD from infancy to adulthood, as well as reveal the pathophysiological commonalities between different forms ILD at the molecular level. The results of these efforts will create large incremental changes in understanding and management of chILD. Since chILD is an umbrella term for a number of conditions most of which imply more than purely medical or scientific expertise, the Action will pay due attention to the larger societal implications of chILD research.
Interstitial lung disease in children (chILD) is a term that describes a collection of more than 200 rare lung disorders. It is a heterogeneous group of non-neoplastic disorders resulting from damage by varying patterns of inflammation and fibrosis with the interstitium as the primary site of injury. As with other orphan diseases, chILD data is lacking on the natural course, phenotypic variability, associations with genotype, and effectiveness of treatments. The disease course is very variable, and depending on more than just the underlying cause; for example, within a given family, the phenotypic variability of chILD such as surfactant protein C mutation is huge. The rarity of individual chILDs contributes to a lack of randomised control trial data on effectiveness of treatments. Management strategies derive from other diseases or are based on physicians experience and remain controversial.
This Action will create a pan-Europe-led network of multidisciplinary clinicians (adult and paediatric), scientists, and patients and their families with the aim of accurate and early diagnosis with structured, potentially personalised, management and therapies. The Action will stimulate and coordinate multidisciplinary research in chILD from infancy to adulthood, as well as reveal the pathophysiological commonalities between different forms ILD at the molecular level. The results of these efforts will create large incremental changes in understanding and management of chILD. Since chILD is an umbrella term for a number of conditions most of which imply more than purely medical or scientific expertise, the Action will pay due attention to the larger societal implications of chILD research
The book ‘Cloud-Based Benchmarking of Medical Image Analysis’ by Allan Hanbury, Henning Müller and Georg Langs is now available at Springer.
This book presents the VISCERAL project benchmarks for analysis and retrieval of 3D medical images (CT and MRI) on a large scale, which used an innovative cloud-based evaluation approach where the image data were stored centrally on a cloud infrastructure and participants placed their programs in virtual machines on the cloud. The book presents the points of view of both the organizers of the VISCERAL benchmarks and the participants.
The book is divided into five parts. Part I presents the cloud-based benchmarking and Evaluation-as-a-Service paradigm that the VISCERAL benchmarks used. Part II focuses on the datasets of medical images annotated with ground truth created in VISCERAL that continue to be available for research. It also covers the practical aspects of obtaining permission to use medical data and manually annotating 3D medical images efficiently and effectively. The VISCERAL benchmarks are described in Part III, including a presentation and analysis of metrics used in evaluation of medical image analysis and search. Lastly, Parts IV and V present reports by some of the participants in the VISCERAL benchmarks, with Part IV devoted to the anatomy benchmarks and Part V to the retrieval benchmark.
This book has two main audiences: the datasets as well as the segmentation and retrieval results are of most interest to medical imaging researchers, while eScience and computational science experts benefit from the insights into using the Evaluation-as-a-Service paradigm for evaluation and benchmarking on huge amounts of data.
The papers ‘Making sense of large datasets without annotations: Analyzing age-related correlations from lung CT scans’ by Yashin Dicente Cid et al. and ‘Convolutional neural networks for an automatic classification of prostate tissue slides with high-grade Gleason score’ by Oscar Jimenez-del-Toro et al. were accepted for the SPIE Medical Imaging 2017 conference.
The conference will take place from the 11th-16th February 2017 in Orlando, Florida, United States
The article “Activity evaluation: a video-based approach to iterative prototype design” by Grace Eden et al. has now been added to the ACM Digital Library
It was originally published in the IHM ’16: Actes de la 28iéme conférence francophone sur l’Interaction Homme-Machine.
Abstract
Naturalistic prototype evaluation is an approach that can be used to assess a technology’s potential practical application in real-world settings and situations (its usefulness) and to evaluate the interface (its usability). This paper discusses our development of a novel prototype evaluation technique; the Activity Evaluation Framework (AEF). AEF is informed by a sociological orientation, Ethnomethodology and uses video in the analysis of participants’ ‘talk-in-interaction’ as they interact with technology. We discuss the use of AEF as a systematic and robust qualitative technique for the evaluation of both the usefulness and usability of prototypes.
Roger Schaer gave a demo presentation of the GoogleGlass and NinaPro projects at the Interaction Homme-Machine 2016 conference.
The conference took place from the 25-28 octobre 2016 at Fribourg, Switzerland.
The article ‘Crowdsourcing Biodiversity Monitoring: How Sharing your Photo Stream can Sustain our Planet‘ by Alexis Joly et al. has been added to the ACM Digital Library.
The article is part of the MM’16: Proceedings of the 2016 ACM on Multimedia Conference.
Abstract
Large scale biodiversity monitoring is essential for sustainable development (earth stewardship). With the recent advances in computer vision, we see the emergence of more and more effective identification tools allowing to set-up large-scale data collection platforms such as the popular Pl@ntNet initiative that allow to reuse interaction data. Although it covers only a fraction of the world flora, this platform is already being used by more than 300K people who produce tens of thousands of validated plant observations each year. This explicitly shared and validated data is only the tip of the iceberg. The real potential relies on the millions of raw image queries submitted by the users of the mobile application for which there is no human validation. People make such requests to get information on a plant along a hike or something they find in their garden but not know anything about. Allowing the exploitation of such contents in a fully automatic way could scale up the world-wide collection of implicit plant observations by several orders of magnitude, which can complement the explicit monitoring efforts. In this paper, we first survey existing automated plant identification systems through a five-year synthesis of the PlantCLEF benchmark and an impact study of the Pl@ntNet platform. We then focus on the implicit monitoring scenario and discuss related research challenges at the frontier of computer science and biodiversity studies. Finally, we discuss the results of a preliminary study focused on implicit monitoring of invasive species in mobile search logs. We show that the results are promising but that there is room for improvement before being able to automatically share implicit observations within international platforms.
The SLDESUTO team (ContextVision/HES-SO) obtained the 3rd place in two tasks of the ‘Tumor Proliferation Assessment Challenge (TUPAC 2016)’. The team used an automated deep learning approach to analyse whole slide digital pathology images.
The corresponding tasks were:
More information can be found here
Sebastian Otalora has joined the Medgift group.
Welcome to him!
An scholarship ‘Bilateral research collaboration with the Asia-Pacific region 2013-2016’ funded by the Swiss State Secretariat for Education, Research and Innovation has been approved for our group.
This scholarship from the Swiss confederation is targeted for projects that colaborate with partners in Malaysia.
The current Microsoft Azure for research award has been increased. The VISCERAL benchmarks as well as other MedGIFT projects are currently sponsored by Azure Research.
For more information regarding the open VISCERAL benchmarks (Anatomy3 continuous, Detection2, Retrieval2) go to www.visceral.eu
The article “Deep learning with convolutional neural networks: a resource for the control of robotic prosthetic hands via electromyography”, by Manfredo Atzori, Matteo Cognolato, Henning Müller, has now been published in Frontiers in Neurorobotics.
ABSTRACT
Natural control methods based on surface electromyography (sEMG) and pattern recognition are promising for hand prosthetics. However, the control robustness offered by scientific research is still not sufficient for many real life applications, and commercial prostheses are capable of offering natural control for only a few movements. In recent years deep learning revolutionized several fields of machine learning, including computer vision and speech recognition. Our objective is to test its methods for natural control of robotic hands via sEMG using a large number of intact subjects and amputees. We tested convolutional networks for the classification of an average of 50 hand movements in 67 intact subjects and 11 transradial amputees. The simple architecture of the neural network allowed to make several tests in order to evaluate the effect of pre-processing, layer architecture, data augmentation and optimization. The classification results are compared with a set of classical classification methods applied on the same datasets. The classification accuracy obtained with convolutional neural networks using the proposed architecture is higher than the average results obtained with the classical classification methods, but lower than the results obtained with the best reference methods in our tests. The results show that convolutional neural networks with a very simple architecture can produce accurate results comparable to the average classical classification methods. They show that several factors (including pre-processing, the architecture of the net and the optimization parameters) can be fundamental for the analysis of sEMG data. Larger networks can achieve higher accuracy on computer vision and object recognition tasks. This fact suggests that it may be interesting to evaluate if larger networks can increase sEMG classification accuracy too.
Prof. Henning Müller will be an invited speaker at the workshop: Challenges in Machine Learning: Gaming and Education (CiML) 2016. This is one of the NIPS workshops and will take place Friday December 9, 2016, Barcelona, Spain.
Motivations
Challenges in machine learning and data science are competitions running over several weeks or months to resolve problems using provided datasets or simulated environments.
The playful nature of challenges naturally attracts students, making challenge a great teaching resource. For this third edition of the CiML workshop at NIPS we want to explore more in depth the opportunities that challenges offer as teaching tools. The workshop will give a large part to discussions around several axes: (1) benefits and limitations of challenges to give students problem-solving skills and teach them best practices in machine learning; (2) challenges and continuous education and up-skilling in the enterprise; (3) design issues to make challenges more effective teaching aids; (3) curricula involving students in challenge design as a means of educating them about rigorous experimental design, reproducible research, and project leadership.
CiML is a forum that brings together workshop organizers, platform providers, and participants to discuss best practices in challenge organization and new methods and application opportunities to design high impact challenges. Following the success of last year’s workshop, in which a fruitful exchange led to many innovations, we propose to reconvene and discuss new opportunities for challenges in education, one of the hottest topics identified in last year’s discussions. We have invited prominent speakers in this field.
We will also reserve time to an open discussion to dig into other topic including open innovation, coopetitions, platform interoperability, and tool mutualisation.
Prof. Adrien Depeursinge was a special guest at the QTIM meeting in Boston last July 27, 2016. The topic of his talk was ‘A 3-D Riesz-Covariance Texture Model for the Prediction of Nodule Recurrence in Lung CT’
Abstract:
We propose a novel imaging biomarker of lung cancer relapse from 3-D texture analysis of CT images. Three-dimensional morphological nodular tissue properties are described in terms of 3-D Riesz-wavelets. The responses of the latter are aggregated within nodular regions by means of feature covariances, which leverage rich intra- and inter-variations of the feature space dimensions. The obtained Riesz-covariance descriptors lie on a manifold governed by Riemannian geometry requiring specific geodesic metrics to locally approximate scalar products. The latter are used to construct a kernel for support vector machines (SVM). The effectiveness of the presented models is evaluated on a dataset of 92 patients with non-small cell lung carcinoma (NSCLC) and cancer recurrence information. Disease recurrence within a timeframe of 12 months could be predicted with an accuracy above 80, and highlighted the importance of covariance-based texture aggregation. At the end of the talk, computer tools will be presented to easily extract 3D radiomics quantitative features from PET-CT images.
The CLEF2016 Working Notes are now on-line on CEUR-WS Proceedings: http://ceur-ws.org/Vol-1609/
The Working Notes correspond to the Conference and Labs of the Evaluation forum that took place in Évora, Portugal, 5-8 September, 2016.