Image processing and machine learning for fully automated probabilistic evaluation of medical images

Šajn, Luka; Kukar, Matja

doi:10.1016/j.cmpb.2010.06.021

Cited by 29 publications

(26 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ML methods have already proven very successful in applications ranging from speech or image recognition 6 , to medical diagnostics 7 , bioinformatics 8 , and economics 9 . There have also been some encouraging examples of using ML to predict biological activities of small molecules 10–12 , solubilities 13 , crystal structures 14 , properties of organic photovoltaics 15 and, recently, compositions of reaction mixtures and/or reaction conditions leading to templated vanadium selenites 16 .…”

Section: Introductionmentioning

confidence: 99%

Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?

Skoraczyński

Dittwald

Miasojedow

et al. 2017

Sci Rep

126

110

View full text Add to dashboard Cite

As machine learning/artificial intelligence algorithms are defeating chess masters and, most recently, GO champions, there is interest – and hope – that they will prove equally useful in assisting chemists in predicting outcomes of organic reactions. This paper demonstrates, however, that the applicability of machine learning to the problems of chemical reactivity over diverse types of chemistries remains limited – in particular, with the currently available chemical descriptors, fundamental mathematical theorems impose upper bounds on the accuracy with which raction yields and times can be predicted. Improving the performance of machine-learning methods calls for the development of fundamentally new chemical descriptors.

show abstract

Section: Introductionmentioning

confidence: 99%

Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?

Skoraczyński

Dittwald

Miasojedow

et al. 2017

Sci Rep

126

110

View full text Add to dashboard Cite

show abstract

“…Five attributes were then processed by Weka (a machine learning based software). [18][19][20][21][22][23][24] In the present study, to achieve the optimization of the results previously found we focused in an ECG analysis using wavelet transforms, statistical information and the original signal processed by LWL, J48, ADTree and RandomForest. These algorithms are inserted in Weka and have already been used in signal processing at different context/applications, as an open source software Weka has a high potential in many other applicantions.…”

Section: Figurementioning

confidence: 99%

Applying learning algorithms to extract anxiety levels using the heart rate variability measure

Magini

Mocaiber

Calembo

et al. 2015

JBGC

View full text Add to dashboard Cite

The classification problems in biological measures have been studied since mathematical methods and statistical tools were created to determine difference between two distinct samples. In this paper we present a mathematical methodology capable of differing 29 non-clinical volunteers with distinct degrees of trait anxiety (high or low) according to the State and Trait Anxiety Inventory (STAI-T) using an electrocardiogram (ECG) data as starting point. Specifically, the wavelet transforms and its statistical measures were used to extract simple patterns from the resting ECG and classify the group as low or high trait anxiety. The Daubechies, Haar and Symlet mother function were used to filter the original ECG data. Then, by means of the Weka Learning Algorithm and using only 5 attributes (Pearson Coefficient from Haar and Symlet, Median from Haar and Mode of Haar and Daubechies) we achieved a higher level of reliability, 96.90% (p < .05), with low training percentages. The results showed the efficiency of this methodology to classify volunteers according to their anxiety levels through an ECG data collection.

show abstract

“…One fast-growing type of application is the use of image processing to improve medical diagnostic effectiveness [2]. Diagnostics that rely on imaging technologies such as Magnetic Resonance Imaging (MRI), Computed Tomography (CT) and X-ray benefit considerably from the practice of image processing.…”

Section: Introductionmentioning

confidence: 99%

“…Diagnostics that rely on imaging technologies such as Magnetic Resonance Imaging (MRI), Computed Tomography (CT) and X-ray benefit considerably from the practice of image processing. Images taken using these clinical procedures can be improved by post-processing algorithms that enhance image quality using contrast enhancement and noise reduction [1], or automatically extract necessary information, as done in image segmentation [1,2]. While most of these tools are implemented as automated detection tools [2], the final decision is always done by an experienced diagnostic physician.…”

Section: Introductionmentioning

confidence: 99%

Improved Diagnostic Process of Multiple Sclerosis Using Automated Detection and Selection Process in Magnetic Resonance Imaging

et al. 2017

View full text Add to dashboard Cite

Abstract:In this paper, we present a new method of displaying Magnetic Resonance (MR) images taken from Multiple Sclerosis (MS) patients. We show that our method can potentially make the diagnostic process far more focused and concise. The method is implemented as an algorithm-based application, which automatically detects MS lesions and reduces the amount of reviewed images by 98% or more. In contrast to existing detection algorithms, our application utilizes five different types of MR images as well as the Digital Imaging and Communications in Medicine (DICOM) standard, supporting a wide range of data sets. After images are selected for file type and relevant brain region, each image is subjected to four separate algorithms, the results of which are combined into a single displayed image for the use of the diagnosing physician.

show abstract

Image processing and machine learning for fully automated probabilistic evaluation of medical images

Cited by 29 publications

References 17 publications

Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?

Predicting the outcomes of organic reactions via machine learning: are current descriptors sufficient?

Applying learning algorithms to extract anxiety levels using the heart rate variability measure

Improved Diagnostic Process of Multiple Sclerosis Using Automated Detection and Selection Process in Magnetic Resonance Imaging

Contact Info

Product

Resources

About