Medical Diagnostic System Research Papers (original) (raw)

Sieci neuronowe są narzędziem o potencjalnie bardzo szerokim zakresie zastosowań. Jednak z praktyki wynika, są one przede wszystkim wykorzystywane w zagadnieniach klasyfikacji sygnałów i obrazów. Z uwagi na swą uniwersalność mogą być też użyte w obszarze diagnostyki. Cechą charakterystyczną tych zastosowań jest to, że te same struktury sieci i podobna metodyka ich uczenia mogą być stosowane do tak odległych na pozór zagadnień jak diagnostyka techniczna i diagnostyka medyczna. Niniejsza praca stanowi próbę opisu jednolitego podejścia do rozwiązywania problemów w zakresie diagnostyki technicznej i diagnostyki medycznej stosowanego przez autorów przy korzystaniu z możliwości stwarzanych przez sztuczne sieci neuronowe.

A Phonocardiogram or PCG is a plot of high fidelity recording of the sounds and murmurs made by the heart with the help of the machine called phonocardiograph. It has developed continuously to perform an important role in the proper and accurate diagnosis of the defects of the heart. As usually with the stethoscope, it requires highly and experienced physicians to read the phonocardiogram. A diagnostic system based on Artificial Neural Networks (ANN) is implemented as a detector and classifier of heart diseases. The output of the system is the classification of the sound as either normal or abnormal, if it is abnormal what type of abnormality is present. In this paper, Based on the extracted time domain and frequency domain features such as energy, mean, variance and Mel Frequency Cepstral Coefficients (MFCC) various heart sound samples are classified using Support Vector Machine (SVM), K Nearest Neighbour (KNN), Bayesian and Gaussian Mixture Model (GMM) Classifiers. The data used in this paper was obtained from Michigan university website.

Nowadays, it is important to design and develop effective applications for improving solution experiences in the fields associated with studies of biomedical and health informatics. Along with rapid developments in computer and electronics technologies , that has become more important in especially twenty-first century. Moving from that, main objective of this study is to deal with the problem of heart sound analysis and disease diagnosis by using a mobile application that can perform the mentioned tasks by having also support from both virtual reality-and augmented reality-oriented components. Thanks to the using features and functions provided over the application, it is possible to analyze signals instantly and have rapid feedback over the interface supported with virtual or mixed reality objects as combining both real and virtual worlds in a common ground. The paper briefly focuses on technical background and essentials of the developed mobile system and then evaluates its performance in the context of different perspectives.

- by Utku Köse
- •
- Artificial Intelligence, Signal Processing, Mobile Technology, Augmented Reality

The detection of double edges in x-ray images of lumbar vertebrae is of prime importance in the assessment of vertebral injury or collapse that may be caused by osteoporosis and other spine pathology. In addition, if the above double-edge detection process is conducted within an automatic framework, it would not only facilitate inexpensive and fast means of obtaining objective morphometric measurements on the spine, but also remove the human subjectivity involved in the morphometric analysis. This paper proposes a novel force-formulation scheme, termed as Pressurized Open Directional Gradient Vector Flow snakes, to discriminate and detect the superior and inferior double edges present in the radiographic images of the lumbar vertebrae. As part of the validation process, this algorithm is applied to a set of 100 lumbar images and the detection results are quantified using analyst-generated ground truth. The promising nature of the detection results bears testimony to the efficacy of the proposed approach

Cardiac surgery is an important medical treatment for coronary vessel patients. Different models have been introduced to determine the risk factors related to side effects of this operation. The goal of this research is to study EuroSCORE (European System for Cardiac Operative Risk Evaluation) as a useful method for predicting the risk of mortality after cardiac surgery, and to introduce a new way of inference, called Fuzzy EuroSCORE. In addition, a systems reliability analysis will be used to calculate the survival possibility of patients after a certain time period after cardiac surgery. To model and simulate the suggested system, eight important parameters of EuroSCORE table are chosen using expert's knowledge and a new method is applied based on a fuzzy inference system. To calculate the risk of mortality after cardiac surgery, the patients are categorized into 3 different groups of low risk, medium risk, and high risk. The range of the mortality risk is determined by appropriate medical data in the fuzzy EuroSCORE system. Additionally, a defect density function for the cardiovascular problem is suggested using the systems reliability analysis. Finally, the prospect of patient's survival after a certain time period after cardiac surgery is predicted.

- by Sina Khanmohammadi and +1
- •
- Fuzzy Logic, Mortality, Healthcare, Cardiac Surgery

Content-Based Image Retrieval (CBIR) locates, retrieves and displays images alike to one given as a query, using a set of features. It demands accessible data in medical archives and from medical equipment, to infer meaning after some processing. A problem similar in some sense to the target image can aid clinicians. CBIR complements text-based retrieval and improves evidence-based diagnosis, administration, teaching, and research in healthcare. It facilitates visual/automatic diagnosis and decision-making in real-time remote consultation/screening, store-and-forward tests, home care assistance and overall patient surveillance. Metrics help comparing visual data and improve diagnostic. Specially designed architectures can benefit from the application scenario. CBIR use calls for file storage standardization, querying procedures, efficient image transmission, realistic databases, global availability, access simplicity, and Internet-based structures. This chapter recommends important and complex aspects required to handle visual content in healthcare.

The aim of this study was to study the diagnostic parameters in
brain tuberculosis (meningeal and parenchymal)
Material and Methods: This study was conducted in the
department of Neurosurgery and Neurology SKIMS for a period
of two years. A total of 61 patients presenting with brain
tuberculosis admitted at skims during these two years were
included in the study.
Results: The most presenting symptom in our study was headache
found in 95.10% followed by vomiting found in 86.90% of
subjects, fever in 78.70%, altered sensorium in 49.20%, seizures in
19.70% and diplopia in18%. Out of 61 patients cranial nerve
involment was found in 34(55.73%) with 11 having more than two
cranial nerves involved. The most common cranial nerve involved
were 3rd and 6th. ADA was positive in 36 of 53 patients of TBM
with a sensitivity of 67.9% and a specificity of 75%. PCR proved
to be highly specific CT scan of head was abnormal in 56 out of
61 patients (91.8%). 12 (19.70%) were in stage I (meningeal
involvement only), 29 (47.50%) were in stage II (parenchymal
involvement only) and 15 (24.60%) were in stage III (both
parenchymal and meningeal invlolvement). The most common
finding in CT head was meningeal enhancement in 43 patients,
hydrocephalus in 37 patients and tuberculomas in 14 patients. The
most common sites of tuberculomas were frontal lobe (n=6;
42.8%), parietal lobe (n=4; 28.5%), followed by cerebellum in 2
patients and occipital in two. Nine patients had single and five
multiple tuberculomas. Of the 14 patients with tuberculomas,
hydrocephalus on CT was seen in 6 patients.
Conclusion: CT scan is a useful diagnostic tool even in very early
stages of TBM. Abnormalities reported on CT scan done are
hydrocephalus, infarcts, basal enhancement, and tuberculomas.
Normal study is reported in up to 20% of the cases.

Keywords: bigdata, data mining, data-driven healthcare, digital hospital, information extraction, machine learning, mhealth, big data analytics

- by Alexandre Prozoroff
- •
- Machine Learning, Data Mining, Information Extraction, Mhealth

This paper proposes a combination of fuzzy standard additive model (SAM) with wavelet features for medical diagnosis. Wavelet transformation is used to reduce the dimension of high-dimensional datasets. This helps to improve the convergence speed of supervised learning process of the fuzzy SAM, which has a heavy computational burden in high-dimensional data. Fuzzy SAM becomes highly capable when deployed with wavelet features. This combination remarkably reduces its computational training burden. The performance of the proposed methodology is examined for two frequently used medical datasets: the lump breast cancer and heart disease. Experiments are deployed with a five-fold cross validation. Results demonstrate the superiority of the proposed method compared to other machine learning methods including probabilistic neural network, support vector machine, fuzzy ARTMAP, and adaptive neuro-fuzzy inference system. Faster convergence but higher accuracy shows a win-win solution of the proposed approach.

We state a problem concerning how to make an effective and proper decision in the presence of data incompleteness. As an example we consider a medical diagnostic system where the problem of missing data is commonly encountered. We propose and evaluate an approach that makes it possible to reduce the influence of missing data on the final result and to improve the quality of the decision. The process involves interval-valued fuzzy set modelling, uncertaintification of classical methods, and finally aggregation of the incomplete results. It was verified that the aggregation results in meaningful and accurate decisions despite the missing data.

- by Andrzej Wójtowicz and +2
- •
- Decision Making Under Uncertainty, Missing Data, Incomplete Information, Aggregation

ABSTRACT A medical knowledge-driven diagnostic process can be supported by AI methods as presented here by an Asperger Syndrome case study. Two methods: consistency-driven pairwise comparisons (CDPC) and automatic understanding (AU) are presented in this study. Deficiencies of a data-driven model for the medical diagnostic process and clinical reasoning are also discussed.

- by Anna Gakamsky
- •
- Mass Spectrometry, Fluorescence, Humans, Animals

Uganda still has shortage of doctors in public health centers, despite the large number of about 400 doctors that graduate from medical schools and universities every year, many of which are willing to work in public hospitals.
Our project therefore therefore is focused towards enabling any available health worker to timely perform diagnosis of patients even in absence of a qualified doctor.

- by timothy kisiga and +2
- •
- Machine Learning, Deep Learning, Medical Diagnostic System

A medical knowledge-driven diagnostic process can be supported by AI methods as presented here by an Asperger Syndrome case study. Two methods: consistency-driven pairwise comparisons (CDPC) and automatic understanding (AU) are presented in this study. Deficiencies of a data-driven model for the medical diagnostic process and clinical reasoning are also discussed.

The chemical nature of the non-tryptophan (non-Trp) fluorescence of porcine and human eye lens proteins was identified by Mass Spectrometry (MS) and Fluorescence Steady-State and Lifetime spectroscopy as post-translational modifications (PTM) of Trp and Arg amino acid residues. Fluorescence intensity profiles measured along the optical axis of human eye lenses with age-related nuclear cataract showed increasing concentration of fluorescent PTM towards the lens centre in accord with the increased optical density in the lens nucleolus. Significant differences between fluorescence lifetimes of “free” Trp derivatives hydroxytryptophan (OH-Trp), N-formylkynurenine (NFK), kynurenine (Kyn), hydroxykynurenine (OH-Kyn) and their residues were observed. Notably, the lifetime constants of these residues in a model peptide were considerably greater than those of their “free” counterparts. Fluorescence of Trp, its derivatives and argpyrimidine (ArgP) can be excited at the red edge of the Trp absorption band which allows normalisation of the emission spectra of these PTMs to the fluorescence intensity of Trp, to determine semi-quantitatively their concentration. We show that the cumulative fraction of OH-Trp, NFK and ArgP emission dominates the total fluorescence spectrum in both emulsified post-surgical human cataract protein samples, as well as in whole lenses and that this correlates strongly with cataract grade and age.

- by Dmitry Gakamsky and +3
- •
- Medical Diagnostic System, Red edge, Tryptophan fluorescence, Post translational modification of proteins

Laboratory and in vivo test results greatly influence further diagnosis and treatment. Therefore, it is essential to select appropriate tests at early stages of the diagnostic process and keep in mind that each test can produce false positive and false negative results. The selection of an optimal test requires a thorough understanding of the terms describing the efficacy of a diagnostic test, such as sensitivity, specificity, positive and negative predictive value, likelihood ratios for a positive and negative test. The present review reports definitions of the relevant terms, along with examples from the allergy practice.

- by Aleksandra Gregorius and +1
- •
- Food Allergy, Allergy, Medical Diagnosis, Medical Diagnostic System

Vesicoureteral reflux (VUR) occurs commonly in children and can cause significant renal damage. The purpose of this study is to assess the changes in renal vasculature with spectral Doppler sonography in patients with VUR. In addition, the possible effects of voiding cystoureterography (VCU) on the kidneys in patients with VUR are investigated by calculating renal resistivity index (RI) values before and after VCU using spectral Doppler sonography. In this prospective study, 114 kidneys of 58 children ages 0 to 16 years were included. The RI values that were calculated before and after VCU and RI values in different grades of VUR were compared statistically. In patients with VUR, the renal parenchymal RI values before and after VCU were significantly higher than those in patients without VUR ( P < .05). The mean pre-VCU RI values were 0.68 ± 0.03 and 0.65 ± 0.05 in patients with VUR and in the control group, respectively, and the mean post-VCU RI values were 0.68 ± 0.03 and 0.65 ...

- by Serhat Avcu
- •
- Resistive index, Clinical Sciences, Medical Diagnostic System, Prospective Study
- by Anna Gakamsky
- •
- Mass Spectrometry, Fluorescence, Humans, Animals

[ENG] This chapter deals with the medical diagnosis from the point of view of abductive reasoning. To this end, proposals such as those known as AKM and GW are analyzed. Throughout the text, the different elements that make up abductive reasoning are discussed, from uncertainty and the prescriptive aspect of abduction to a case analysis such as psychiatry. The final reflections on the imbrication between fact and value link medical diagnosis with ethics and the cognitive dimension, thus forming an integral vision of abductive reasoning.

The analysis of physiological data plays a significant role in medical diagnostics. While state-of-the-art machine learning models demonstrate high levels of performance in classifying physiological data clinicians are slow to adopt them. A contributing factor to the slow rate of adoption is the "black-box" nature of the underlying model whereby the clinician is presented with a prediction result, but the rationale for that result omitted or not presented in an interpretable manner. This gives rise to the need for interpretable machine learning models such that clinicians can verify, and rationalise, the predictions made by a model. If a clinician understands why a model makes a prediction, they will be more inclined to accept a models assistance in analysing physiological data. This paper discusses some of the latest findings in interpretable machine learning. Thereafter, based on these findings, three models are selected and implemented to analyse ECG data that are both accurate and exhibit a high level of interpretability.