Wyniki wyszukiwania - BazTech

1

A progressive and cross-domain deep transfer learning framework for wrist fracture detection

Karam Christophe, El Zini Julia, Awad Mariette, Saade Charbel, Naffaa Lena, El Amine Mohammad

Journal of Artificial Intelligence and Soft Computing Research

|

2022

|

Vol. 12, No. 2

101--120

EN

There has been an amplified focus on and benefit from the adoption of artificial intelligence (AI) in medical imaging applications. However, deep learning approaches involve training with massive amounts of annotated data in order to guarantee generalization and achieve high accuracies. Gathering and annotating large sets of training images require expertise which is both expensive and time-consuming, especially in the medical field. Furthermore, in health care systems where mistakes can have catastrophic consequences, there is a general mistrust in the black-box aspect of AI models. In this work, we focus on improving the performance of medical imaging applications when limited data is available while focusing on the interpretability aspect of the proposed AI model. This is achieved by employing a novel transfer learning framework, progressive transfer learning, an automated annotation technique and a correlation analysis experiment on the learned representations. Progressive transfer learning helps jump-start the training of deep neural networks while improving the performance by gradually transferring knowledge from two source tasks into the target task. It is empirically tested on the wrist fracture detection application by first training a general radiology network RadiNet and using its weights to initialize RadiNetwrist, that is trained on wrist images to detect fractures. Experiments show that RadiNetwrist achieves an accuracy of 87% and an AUC ROC of 94% as opposed to 83% and 92% when it is pre-trained on the ImageNet dataset. This improvement in performance is investigated within an explainable AI framework. More concretely, the learned deep representations of RadiNetwrist are compared to those learned by the baseline model by conducting a correlation analysis experiment. The results show that, when transfer learning is gradually applied, some features are learned earlier in the network. Moreover, the deep layers in the progressive transfer learning framework are shown to encode features that are not encountered when traditional transfer learning techniques are applied. In addition to the empirical results, a clinical study is conducted and the performance of RadiNetwrist is compared to that of an expert radiologist. We found that RadiNetwrist exhibited similar performance to that of radiologists with more than 20 years of experience. This motivates follow-up research to train on more data to feasibly surpass radiologists’ performance, and investigate the interpretability of AI models in the healthcare domain where the decision-making process needs to be credible and transparent.

2

Wykorzystanie medycznych baz danych w praktyce klinicznej fizjoterapeuty

Mikołajewski Dariusz, Mikołajewska Emilia

Studia i Materiały Informatyki Stosowanej

|

2018

|

nr 2

29--34

PL

Medyczne bazy danych umożliwiają systematyczne gromadzenie, analizę i interpretację wyników badań klinicznych, stanowiąc uzupełnienie wiedzy, doświadczenia i intuicji wszystkich członków zespołu wielospecjalistycznego, w tym fizjoterapeutów. Promuje to twórcze dyskusje w zespole i sprzyja tworzeniu lokalnych autorytetów klinicznych, oszczędza czas i przyśpiesza rozwój praktyki klinicznej oraz badań. Bezpośrednią korzyścią dla pacjenta jest wybór najefektywniejszej z dostępnych terapii. Wpisuje się to czytelnie w paradygmat Medycyny Opartej na Faktach (ang. Evidence Based Medicine – EBM). Praktyczna implementacja takiego podejścia wymaga właściwego przygotowania kadr medycznych, zarówno w ramach kształcenia dyplomowego, jak i podyplomowego oraz dalszego nabywania doświadczenia w ramach praktyki klinicznej i samokształcenia. Internet w tym zakresie niweluje znaczenie odległości, a bariera językowa ma coraz mniejsze znaczenie

EN

Medical data bases allow for systematic gathering, analysis and interpretation clinical research’ outcomes. They can be useful supplement of knowledge, experience and intuition of all members of therapeutic multidisciplinary team, inculding physical therapists. This approach promotes creative discussions within the team and creating of local clinical authorities, saves time and advances development of clinical practice and research. Direct advantage for the patient is the choice of the most effective accessible therapy. This process is a part of Evidence Based Medicine (EBM) paradigm. Practical implementation of the EBM-based approach requires correct preparation of medical staff, including graduate and postgraduate education, further experience and self-education. In the area of EBM use of internet levels significance of distance and language.

3

Qualitative evaluation of distributed clinical systems supporting research teams working on large-scale data

Dziubich T.

TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk

|

2017

|

Vol. 21, No 3

215--231

EN

In this paper, five contemporary scalable systems to support medical research teams are presented. Their functionalities extend from heterogeneous unstructured data acquisition through large-scale data storing, to on-the-fly analyzing by using robust methods. Such kinds of systems can be useful in the development of new medical procedures and recommendation rules for decision support systems. A short description of each of them is provided. Further, a set of the most important features is selected, and a comparison based-on it is performed. The need for high performance computing is emphasized. A general discussion how to improve the existing solutions or develop new ones in the future is also presented.

4

Cardiac arrhythmia alarm from optical interferometric signals during resting or sleeping for early intervention

Cikajlo I., Šprager S., Erjavec T., Zazula D.

Biocybernetics and Biomedical Engineering

|

2016

|

Vol. 36, no. 1

267--275

EN

Diagnostics of cardiac arrhythmias and frequent interventions may contribute to early detection of diseases or even prevent sudden death. Generally electrocardiograph with several on body electrodes at outpatient clinic is applied and the procedure requires a medical expert. We propose cardiac arrhythmia estimation on the basis of heartbeat detection with optical fibers integrated in the bedding. The modified Michelson's interferometer with error detection was used to measure and maximum a-posteriori probability was used to estimate the beat-to-beat intervals. The consistency of heartbeat intervals was examined with simultaneous measurement with clinical electrocardiograph in 10 healthy volunteers and 10 patients with diagnosed heart arrhythmias. Heart beat interval data obtained in patients were examined and irregularities/arrhythmias were identified from the medical guidelines. The current system enables assessment also in home environment without any on-body sensor placement or required assistance. Thus early intervention is possible as the irregularities are submitted to the nurse on duty and stored in the database for subsequent more detailed analysis.

5

Classification techniques for non-invasive recognition of liver fibrosis stage

Krawczyk B., Woźniak M., Orczyk T., Porwik P., Musialik J., Błońska-Fajfrowska B.

Journal of Medical Informatics & Technologies

|

2012

|

Vol. 20

121--127

EN

Contemporary medicine should provide high quality diagnostic services while at the same time remaining as comfortable as possible for a patient. Therefore novel non-invasive disease recognition methods are becoming one of the key issues in the health services domain. Analysis of data from such examinations opens an interdisciplinary bridge between the medical research and artificial intelligence. The paper presents application of machine learning techniques to biomedical data coming from indirect examination method of the liver fibrosis stage. Presented approach is based on a common set of non-invasive blood test results. The performance of four different compound machine learning algorithms, namely Bagging, Boosting, Random Forest and Random Subspaces, is examined and grid search method is used to find the best setting of their parameters. Extensive experimental investigations, carried out on a dataset collected by authors, show that automatic methods achieve a satisfactory level of the fibrosis level recognition and may be used as a real-time medical decision support system for this task.

6

Hepatic liver diseases - methods for diagnosis and medical informatics for treatment support

Tsakona A., Paschali K., Tsolis D., Skapetis G.

Journal of Medical Informatics & Technologies

|

2011

|

Vol. 17

211--218

EN

Liver diseases and more specifically viral hepatitis are at the center of interest due to their global spreading, even in the most developed countries. The range of symptoms, the complications and the course of the disease have imposed the operation of liver centers at the outpatients' departments of hospitals, where the contribution of several specialized doctors the disease is diagnosed, prevented and treated. Many patients suffer from hepatitis without knowing it either because they manifest no symptoms or because the infection is not traced through the usual lab tests. This paper focuses on studying and proving how the systematic reading of the main liver diseases and the methods through which the doctor makes the diagnosis can help the study and analysis of a series of steps that have to be followed in order to treat the disease. Then, the use of a modern information system using Medical Informatics technologies is proposed so as both the task of diagnosis and the efforts to treat and overcome the problems related to the liver disease to be supported.

7

Informatics in the service of visually disabled children

Koziak L.

Czasopismo Techniczne. Mechanika

|

2008

|

R. 105, z. 3-M

121-128

EN

The progress in modern technologies in the area of information and communication has an influence on institutions and individuals. At schools, informatics may be helpful to blind children in the form of numerous teaching aids. For blind children, the computer is an excellent didactic aid which stimulates their development. Computer programs create a natural potential for individualization of the learning process, diversified depending on the child's predispositions and degree of development, and the type and degree of sight damage.

PL

Postęp nowoczesnych technologii w obszarze informatyki i komunikacji ma wpływ na instytucje i jednostki. W szkole informatyka może być pomocna niewidomym dzieciom poprzez różnego rodzaju środki dydaktyczne. Dla niewidomych dzieci komputer jest doskonałym środkiem dydaktycznym, który stymuluje ich rozwój. Programy komputerowe tworzą naturalne możliwości dla indywidualizacji procesu nauczania, dostosowują zakres do poziomu rozwoju dziecka i stopnia utraty wzroku.

8

Metody zwiększające precyzyjność wyszukiwania informacji-automatyczna kategoryzacja

Kozak K.

Studia Informatica

|

2004

|

Vol. 25, nr 3

71-81

PL

Niniejszy artykuł poświęcony jest metodom organizacji dokumentów uzyskanych w wyniku wyszukiwania i ich zastosowania w aplikacjach. Opisana metoda polega na automatycznej kategoryzacji dokumentów,a przykład zastsowania tej metodyw systemach zawierających w swoim repozytorium dokumenty pochodzi z dziedziny medycyny - psychiatrii.

EN

In this article arę described methods and their implementation to the systems for organize documents from search results. Below is described approach that automatically categorize documents. This approach was implemented to systems which contain and arę repository for scientific documents from domain medicine -psychiatry.

9

Electronic patient record, expert system and mobile monitoring in Cardio.net project

Rudowski R., Grabowski M., Sierdziński J., Guć M.

Biocybernetics and Biomedical Engineering

|

2004

|

Vol. 24, no. 1

61-71

EN

Cardio.net project is aimed at the implementation of the prototype system of telecardiology in Mazovia and Pomerania Regions of Poland. It is planned as a tool to reduce the time from the onset of symptoms to the cardiological intervention, which can significantly reduce the mortality of patients with the acute coronary syndromes (ACS). The system will enable teleconsultations, from the reference centers to the regional hospitals, regarding the need for invasive cardiological intervention. If there is such a need then the patient will be transported to the reference center. The Cardio.net tasks include the design of telecommunication infrastructure, incorporating data transmission standards HL7, DICOM, work on the security and safety of the system, the design of Electronic Patient Record (EPR), uniform archive of EPR's and also the expert system. The EPR consist of two types of forms: a) personal data forms, b) forms supporting the physician work (case history, diagnosis, physical examination, the clinical course of hospitalization, additional tests and drugs). The tools used for creating EPR and archive of EPR's were PHP, Java Script, XML and MySQL. The risk stratification and therapeutic decision making in ACS are the main tasks of the expert system. There are two main therapeutic strategies: early invasive versus conservative. The chosen strategy depends on several factors. The expert system has been verified on the archival data of 147 patients with ACS. In 127 (86%) the expert system suggested the same strategy as applied by the cardiologist. In conclusion: uniform EPR and archive of EPR's give a chance of improving the quality of the cardiological care due to teleconsultations performed directly by specialists or indirectly by the expert system. Mobile monitoring and alerting can decrease the time from the onset of symptoms to the therapeutic intervention thus improving the patients safety,

10

The teaching of medical informatics to medical students including a retrospective evaluation

Lane V., Whitehead S.

Journal of Medical Informatics & Technologies

|

2002

|

Vol. 3

MI109--117

EN

It is commonly suggested that (i) Information and Communication Technology (ICT) is essential for the delivery of quality healthcare [17], and (ii) the role of doctors is crucial in the implementation of ICT-based systems in hospitals and in general practice. Based on these two premises, this paper argues for a medical informatics course for medical students. The paper outlines the content of a one-semester elective in medical informatics that was presented over eight academic years. The course pre-dates the recommendations of the International Medical Informatics Association, IMIA [7]. Nevertheless, it is compatible with the recommendations of IMIA and also with the earlier findings of the EDUCTRA study [5]. The principles formulated in this paper are presented in order to help other educationalists (i) to obtain insights and (ii) thence to assist them to engineer medical informatics changes into their medical school environments. Each time that the course was presented, participants were asked for their evaluation of the course [9]. These views were favourable. It was thought that a follow-up survey of these same respondents - now practising doctors, some of whom took the course 10 years ago - might give further insights into the relevance and value of this course. The views of these practising doctors are presented within the paper.

11

Clinical databases originating in electronic patient records

Zvarova J.

Biocybernetics and Biomedical Engineering

|

2002

|

Vol. 22, no. 1

43-60

EN

The paper is focusing on new technologies to standardize electronic documentation in Czech healthcare environment. Various approaches for collecting and storing biomedical information using new information technologies are considered with respect to all legislative and standarization norms in the Czech Republic. Development of new approaches is based on experience gathered in the European 14C-TripleC project. New approaches for implementation of European ideas in the Czech healthcare environment are given. The focus on Czech healthcare record is presented. Special features of electronic healthcare record to serve in telematic applications for exploring electronic medical guidelines and medical decision support systems are discussed.

12

A modern technologies in medical expert systems

Kącki E., Małolepszy A., Stempczyńska J.

Journal of Applied Computer Science

|

2001

|

Vol. 9, nr 1

23-32

EN

The paper presents organization and performance of the module structure expert system for medicine and discussion on certain information technology problems connected with building and exploitation of the system. The considered problems mainly deal with knowledge acquisition automation and hypermedia applications for well communication. After discussion some remarks on conception of knowledge acquisition and dialog organization as well as some conclutions relative to their implementability and utilizability are also given.

13

The project of expert system aided decision process in medicine

Puchała E.

Journal of Medical Informatics & Technologies

|

2000

|

Vol. 1

CS 65--70

EN

The paper deals with a concept of the expert system for family doctor practices. This system will be realized by Division of Systems and Computer Networks Wroclaw University of Technology in co-operation with Medical Academy of Wroclaw. The family doctor loads data (by user interface) to the system, which describes patients state of health. By this way we can obtain, so-called feature vector. Data in feature vector, may be complete by information collecting from different type of diagnosis equipment like electrocardiograph, Rontgen unit, ultrasonic unit and so on. Number of features may be optional, but for investigation goals only 25 features were used. The user can load to the system less then 25 features. It is important in case then we have problems with results of medical examination extraction. On the basis of a patient information, system gives us decisions which aided medical treatment. Information from data base (set of feature vectors) is processed by expert system. It leads to so called “statistical data base” establishment. System consists of three functionally connected parts: 1. data collecting and data processing, 2. decision part, 3. user’s interface. Expert system can give us unlimited number of decision. In the decision expert system, adaptive algorithms are used. It means that in case of some features are not available, expert system starts algorithms which these feature estimates (algorithms take into account whole patients population’s features and typical for individual patient features).

PL

Celem pracy jest przedstawienie opracowywanego w Zakładzie Systemów i Sieci Komputerowych Politechniki Wrocławskiej projektu systemu ekspertowego wspomagającego podejmowanie decyzji lekarskich. W założeniu, system funkcjonuje w oparciu o „statystyczną” bazę wiedzy współpracującą z oryginalnymi algorytmami wnioskowania.

14

Object and component oriented modelling of complex medical systems

Filutowicz J., Filutowicz Z., Szczepaniak P., Kmiecik A.

Journal of Applied Computer Science

|

1999

|

Vol. 7, nr 1

91-98

EN

The paper describes modelling of a complex information system supporting the work of a dialysis ward and scientific research carried out there. The system is designed to process multimedia information on patients, health care staff, and management of medical material and equipment. This information proves very useful for scientific research and supports physicians in taking decisions concerning extracorporeal dialysis equipment control. The decision-taking system together with optimal control algorithms are used to assess adequacy of dialyses, which results in higher efficiency of dialyses and improvement of life quality of dialysed patients. The new generation of information systems needs making use of scattered multimedia information, algorithms of computer intelligence and user-friendly communication. To meet these requirements it becomes necessary to apply environments enabling modelling of customer's requirements, information system structure and computer-man communication interface. The technology of component and object oriented modelling allows multiple code usage, and ensures higher reliability and easier modification of the system.

15

Multimedia elements in the PHARMACY2 system

Stempczyńska J., Kącki E., Misiak M.

Journal of Applied Computer Science

|

1999

|

Vol. 7, nr 1

55-62

EN

The paper describes the PHARMACY2 system being a modification of PHARMACY1 worked out earlier by the authors and implemented at the Chemotherapy Clinic of the Medical University of Łódź. The main change lies in the introduction of multimedia to the system, which improves communication capabilities of the system and enables better assistance of medical education. In the system a special emphasis is put on cytostatic drugs.