An adaptive multiscale AM-FM texture analysis system with application to hysteroscopy imaging

Constantinou, Ioannis; Pattichis, Marios; Tanos, V.; Neofytou, Marios; Pattichis, Constantinos S.

doi:10.1109/bibe.2012.6399760

Cited by 4 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The introduction of a multiscale approach in ) demonstrated that the method can be used to reconstruct general images. In particular, a multi-scale AM-FM representation led to the important application of population screening for diabetic retinopathy 6 as documented in (Agurto et al, 2010), (Rahim et al, 2015, hysteroscopy image assessment (Constantinou et al, 2012), fMRI and MRI image analysis (Loizou et al, 2011b), and atherosclerotic plaque ultrasound image and video analysis (Loizou et al, 2011a). Alternatively, the definition of multidimensional AM-FM transforms over curvilin-6.…”

Section: Image Analysis Using Multiscale Am-fm Image Decompositionsmentioning

confidence: 99%

What do we need to build explainable AI systems for the medical domain?

Holzinger,

Biemann,

Pattichis

et al. 2017

Preprint

148

145

View full text Add to dashboard Cite

Artificial intelligence (AI) generally and machine learning (ML) specifically demonstrate impressive practical success in many different application domains, e.g. in autonomous driving, speech recognition, or recommender systems. Deep learning approaches, trained on extremely large data sets or using reinforcement learning methods have even exceeded human performance in visual tasks, particularly on playing games such as Atari, or mastering the game of Go. Even in the medical domain there are remarkable results. However, the central problem of such models is that they are regarded as black-box models and even if we understand the underlying mathematical principles of such models they lack an explicit declarative knowledge representation, hence have difficulty in generating the underlying explanatory structures. This calls for systems enabling to make decisions transparent, understandable and explainable. A huge motivation for our approach are rising legal and privacy aspects. The new European General Data Protection Regulation (GDPR and ISO/IEC 27001) entering into force on May 25th 2018, will make black-box approaches difficult to use in business. This does not imply a ban on automatic learning approaches or an obligation to explain everything all the time, however, there must be a possibility to make the results re-traceable on demand. This is beneficial, e.g. for general understanding, for teaching, for learning, for research, and it can be helpful in court. In this paper we outline some of our research topics in the context of the relatively new area of explainable-AI with a focus on the application in medicine, which is a very special domain. This is due to the fact that medical professionals are working mostly with distributed heterogeneous and complex sources of data. In this paper we concentrate on three sources: images, *omics data and text. We argue that research in explainable-AI would generally help to facilitate the implementation of AI/ML in the medical domain, and specifically help to facilitate transparency and trust.

show abstract

Section: Image Analysis Using Multiscale Am-fm Image Decompositionsmentioning

confidence: 99%

What do we need to build explainable AI systems for the medical domain?

Holzinger,

Biemann,

Pattichis

et al. 2017

Preprint

148

145

View full text Add to dashboard Cite

show abstract

“…In conclusion, significant research work is still needed in several directions to enable the deployment of CAD hysteroscopic imaging systems in successful clinical practice. More specifically, there is a need to: 1) further enhance the proposed standardized CAD system with radial distortion correction, 2) investigate the performance of ROI image multiscale feature analysis like amplitude modulation-frequency modulation (AM-FM) features that showed promising results in hysteroscopic imaging [26] as well as color wavelet features that were used successfully in endoscopic tumor detection [27], 3) incorporating navigation functionality for suspicious tissue localization and identification [28], and 4) evaluation on more subjects and in more clinics.…”

Section: Discussionmentioning

confidence: 99%

“…Hysteroscopy imaging CAD protocol covering image calibration and color correction, video acquisition, texture analysis and classification, and diagnosis. This protocol was updated from previously published protocols by our group in [5], [7], and [26]. suspected endometrial lesions, and 2) 22 patients with normal endometrium.…”

Section: A Datasetmentioning

confidence: 99%

Computer-Aided Diagnosis in Hysteroscopic Imaging

Neofytou

Tanos

Constantinou

et al. 2015

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

The paper presents the development of a computer-aided diagnostic (CAD) system for the early detection of endometrial cancer. The proposed CAD system supports reproducibility through texture feature standardization, standardized multifeature selection, and provides physicians with comparative distributions of the extracted texture features. The CAD system was validated using 516 regions of interest (ROIs) extracted from 52 subjects. The ROIs were equally distributed among normal and abnormal cases. To support reproducibility, the RGB images were first gamma corrected and then converted into HSV and YCrCb. From each channel of the gamma-corrected YCrCb, HSV, and RGB color systems, we extracted the following texture features: 1) statistical features (SFs), 2) spatial gray-level dependence matrices (SGLDM), and 3) gray-level difference statistics (GLDS). The texture features were then used as inputs with support vector machines (SVMs) and the probabilistic neural network (PNN) classifiers. After accounting for multiple comparisons, texture features extracted from abnormal ROIs were found to be significantly different than texture features extracted from normal ROIs. Compared to texture features extracted from normal ROIs, abnormal ROIs were characterized by lower image intensity, while variance, entropy, and contrast gave higher values. In terms of ROI classification, the best results were achieved by using SF and GLDS features with an SVM classifier. For this combination, the proposed CAD system achieved an 81% correct classification rate.

show abstract

“…The introduction of a multiscale approach in [1] demonstrated that the method can be used to reconstruct general images. In particular, a multi-scale AM-FM representation led to the important application of population screening for diabetic retinopathy as documented in [7], hysteroscopy image assesment [8], fMRI and MRI image analysis [9], and atherosclerotic plaque ultrasound image and video analysis [10]. Alternatively, the definition of multidimensional AM-FM transforms over curvilinear coordinate systems has led to the earlier development of very low bitrate video coding as demonstrated in [11], [12].…”

Section: Introductionmentioning

confidence: 99%

Multiscale AM-FM image reconstructions based on elastic net regression and Gabor filterbanks

Constantinou

Pattichis

2013

2013 Asilomar Conference on Signals, Systems and Computers

Self Cite

View full text Add to dashboard Cite

The paper proposes the use of elastic net regression for reconstructing images from AM-FM components. Current AM-FM reconstruction methods are based on Dominant Component Analysis (DCA), multi-scale DCA, and Channel Component Analysis (CCA). The paper introduce a variation on CCA that uses elastic net regression to minimize the number of channels that are used in the reconstruction. The new approach is validated using a family of Gabor filterbanks that is parameterized by an overlap index. The results show that the elastic net regression component selection algorithm performs significantly better than multiscale DCA.

show abstract

An adaptive multiscale AM-FM texture analysis system with application to hysteroscopy imaging

Cited by 4 publications

References 14 publications

What do we need to build explainable AI systems for the medical domain?

What do we need to build explainable AI systems for the medical domain?

Computer-Aided Diagnosis in Hysteroscopic Imaging

Multiscale AM-FM image reconstructions based on elastic net regression and Gabor filterbanks

Contact Info

Product

Resources

About