Bayesian inference framework for bounded generalized Gaussian‐based mixture model and its application to biomedical images classification

Abstract

Biomedical image classification problem has attracted a lot of attention in medical engineering community and medicine applications. Accurate and automatic classification (eg, normal/abnormal or malignant/benign) has a variety of applications such as automatic decision making and is known to be very challenging. In this research, we address this problem by investigating the effectiveness of Bayesian inference methods for statistical bounded mixture models. Indeed, a novel approach termed as Bayesian learning for bounded generalized Gaussian mixture models is developed. The consideration of bounded mixture models is encouraged by their capability to take into account the nature of the data that is compactly supported. Furthermore, the consideration of Bayesian inference is more attractive compared to frequentist reasoning. In this work, we address main issues related to accurate data classification such as the effective estimation of the model's parameters and the selection of the optimal model complexity. Moreover, the problem of over-or under-fitting is treated by taking into account the uncertainty through introducing prior information about the model's parameters. A comparative study between different Gaussian-based models is also performed to evaluate the performance of the proposed framework. Experiments have been conducted on challenging biomedical image datasets that involve retinal images for diabetic retinopathy detection and mammograms for breast cancer detection. Obtained results are encouraging and show the benefits of our Bayesian framework. K E Y W O R D SBayesian inference, biomedical imaging, bounded mixture models, generalized Gaussian distribution, image classification, Markov chain Monte Carlo (MCMC)

Bayesian Learning of Shifted-Scaled Dirichlet Mixture Models and Its Application to Early COVID-19 Detection in Chest X-ray Images

“…Therefore, a better solution especially for our case (i.e., when dealing with complex medical noisy data including COVID-19 infection) is to develop a more robust alternative based on fully Bayesian inference approach. We recall that Bayesian estimation has attracted a lot of attention for many applications [ 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 ]. It is also known that the Bayesian approach may be more practical due to the existance of powerful simulation techniques like MCMC [ 29 ].…”

Discriminative Learning Approach Based on Flexible Mixture Model for Medical Data Categorization and Recognition

“…M-step : Update all parameters for the model. Θ j using Equation (5). end Allocate all vectors to their appropriate clusters using maximum posterior probability (MAP) estimate.…”

“…Nowadays, the trend of data mining applications in healthcare is remarkable as there are huge datasets in this sector that require a deep analysis by effective model-based techniques derived from artificial intelligence and machine learning areas. Large scale artificial intelligence and machine learning tools are increasingly successful in image-based diagnosis and have been employed for medical decision making [ 1 , 2 , 3 , 4 , 5 ] and other complex problems like scene and web pages categorization [ 6 , 7 , 8 ], retinal images classification [ 9 ] and action recognition [ 10 ]. The manual processing of these tasks is difficult, tedious and time consuming and so it is important to move to automatic methods, which are able to learn models from labeled and non-labeled data and allow faster and more accurate decisions.…”

Markov Chain Monte Carlo-Based Bayesian Inference for Learning Finite and Infinite Inverted Beta-Liouville Mixture Models

“…It is noted that, only few works have investigated the Liouville distributions in the past such as with Beta-Liouville (IBL) mixture in [4], hence more in-depth exploitation with these distributions is needed to directly and accurately model data. Accordingly, in this paper, we are first motivated by developing a mixture model based on IBL, then we are interested by the promising results obtained recently by Bayesian inference approaches such as with Markov chain Monte Carlo (MCMC) and Metropolis-Hasting algorithms to learn various mixture models [16]- [19]. MCMC is one of the most powerful Bayesian learning sampling algorithms that outperforms deterministic methods (for more details on MCMC the reader could refer to [17]).…”