Deep learning and handcrafted feature based approaches for automatic detection of angiectasia

Pogorelov, Konstantin; Ostroukhova, Olga; Petlund, Andreas; Halvorsen, Pål; Lange, Thomas de; Espeland, Håvard; Kupka, Tomas; Griwodz, Carsten; Riegler, Michael

doi:10.1109/bhi.2018.8333444

Cited by 18 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using the angiecstasia segmentation public dataset [23], we showed that the detection and the localization subsystems of DeepEIR can reach outstanding performance that exceeds clinical requirements (sensitivity and specificity higher than 85%). In summary, we achieved a sensitivity of 88% and a specificity of 99.9% for pixel-wise angiectasia localization, and a sensitivity of 98% and a specificity of 100% for frame-wise angiectasia detection [93].…”

Section: Contributionsmentioning

confidence: 78%

“…Finally, we implemented universal GAN-based localization-via-segmentation and detection-via-localization modules, which allowed us to achieve both frame-and pixel-wise high-precision polyp detection and localization [92]. We later extended this approach to bleeding [129] and angiectasia [93] lesions, which resulted in outstanding detection and localization performance, which is to our best knowledge, better than the state-of-the-art in angiectasia detection and localization.…”

Section: Discussionmentioning

confidence: 99%

“…In case of already pre-trained an DCNN, the deep feature vector contain information about all the image's high-level features in a compact form. For the used architectures, the size of the vector with deep features is pre-defined [93] and it does neither depend on its single-or multi-class nature, nor on the number of classes supported by the specific DCNN. In contrast, deep concept is an output of the top layer of the multiclass classification DCNN.…”

Section: Deep-feature-based Detectionmentioning

confidence: 99%

“…We shared our experience regarding how multimedia researchers can apply their knowledge in the medical field and published the article in the ACM multimedia Brave New Idea track [116]. In addition to the DeepEIR system [25,26,61,62,63,64,80,87,90,91,92,93,94,95,96,97,98,99,100,101,102,115,117,118,119,120,121,129] and side applications of its subsystems [13,14,15,16,55,84,113,122], this can be seen as an important contribution of this thesis to the research society.…”

Section: Summary and Contributionsmentioning

confidence: 99%

“…In order to extend the lesion detection capabilities of the DeepEIR system, we researched and developed a GAN-based detection and localization approach for the angiectasia GI tract lesion [93]. Also, inspired by the great success of our angiectasia detection approach, we researched and developed a GAN-based polyp detection and localization approach [92].…”

Section: Summary and Contributionsmentioning

confidence: 99%

See 4 more Smart Citations

Lire

Lux

Riegler

Halvorsen

et al. 2016

Proceedings of the 7th International Conference on Multimedia Systems

Self Cite

View full text Add to dashboard Cite

Advanced and automated medical systems have been in the research focus for a long time. Together with the rapid development of sensing devices, the modern information analysis methods allow the new wave of computer-assisted systems to improve health care, quality of life, and patient survival rate. Together with the traditional computer vision and medical imaging, core competencies of the multimedia community such as integration and analysis of data from several sources, real-time processing and the assessment of usefulness for end-users play an essential role for the successful improvement of health care systems addressing challenges and open problems in the field of medicine. Our work explores different fields in multimedia research, starting from collection and annotation of multimedia data through automatic analysis of content and efficient processing of workloads to visualization and results representation. We have researched and developed a holistic medical multimedia system addressing a use case with an important medical and societal impact. We target lesions and findings detection and localization in the gastrointestinal (GI) tract of the human body in order to be able to support medical experts in their daily routine work. The early and precise detection of abnormalities in the GI tract greatly increases the chance of successful treatment if the initial observation of disease indicators occurs before the patient notices any symptoms, it is a non-trivial task that can be, however, efficiently automated. We investigated the GI tract visual analysis from a multimedia research point of view via several steps of research and development. First, we looked into the problem of medical data acquisition. We collected, annotated, and published several datasets and data annotation tools as open source. Then, we designed and developed a set of lesion and findings detection and localization approaches based on hand-crafted methods as well as on global-, local-and deepfeature-based methods, which serves as the algorithmic basis of our system. Next, we created a holistic medical multimedia system called DeepEIR. We researched and developed different subsystems for our DeepEIR system, namely (i) the data exploration and annotation subsystem, which makes it possible to collect and annotate data and transfer knowledge from medical experts into our system; (ii) the detection and localization subsystem, which perform medical data analysis in order to detect and localize lesions and findings; and (iii) the visualization and results representation subsystem that provides the information to medical personnel. Furthermore, the focus of the DeepEIR system lies on the accurate and time-efficient processing of multimedia data. We investigated, therefore, parallel and distributed processing, GPU-based acceleration and different classification and segmentation approaches that are evaluated and compared with state-of-the-art methods, algorithms, and systems. We demonstrated that the DeepEIR system could outperform state-of-the-art app...

show abstract

Section: Contributionsmentioning

confidence: 78%