International journal of bio-medical computing最新文献

This paper describes a comprehensive approach for the assessment of the impact of (partial) Picture Archiving and Communication Systems (PACS). The approach is developed, based on actual clinical experience in three European hospitals and tested in these environments. The approach departs from a thorough analysis of the working procedures and information flows before implementation, both descriptive and quantitative. On the basis of this analysis, quantitative (and hence testable) objectives of the implementation are defined. The implementation strategy is defined after comparison of various scenarios, taking costs and effects for both the final and the transition phases into account. The approach is supported by a comprehensive evaluation protocol and a software package (PACER). The approach is demonstrated in this paper by applying it on a hypothetical PACS implementation for CT, ultrasound and for the part of the radiology department serving ICU. The objectives of this PACS are:

• (1)

  — to shorten the turn around time between the radiology department and ICU from 4 h to 30 min

• (2)

  — to save 2000 m² of film per year and

• (3)

  — to save personnel time.

In this case the PACS is introduced in three phases and completed after three years. The cost analysis shows that, if started in 1995, a financial break even point is reached after 6 years, when comparing costs for the film-based system with those of the PACS.

Experiences in the three sites show that the approach helps to harvest potential benefits, allowing a cost-effective implementation of PACS.

The Unified Medical Language System (UMLS) contains and organizes a large number of terms from a variety of biomedical terminology systems. This study examines the relevance of the UMLS content and structures to the specific purpose of the conceptual representation of medical procedures. The MAOUSSC modelling is a compositional formalism with a description of elementary procedures in terms of elementary concept entities and combinations of such descriptions into more complex ones. The UMLS knowledge base is expected to provide semantically categorized medical concepts and interconcept relations. A method to reuse the UMLS has been developed. Quantitative and qualitative results are presented. Some difficulties in reusing the UMLS as a background knowledge are related to the preeminence of some terminology sources and to the instanciation of interconcept links. Other ones suggest that purpose-independence in categorization cannot be achieved.

This article describes the application of Multi-Layer Perceptron (MLP), Probabilistic Neural Network and Kohonen's Learning Vector Quantization to the problem of diagnosing Multiple Sclerosis. The classification information is obtained from brainstem trigeminal evoked potential. The performance of the neural networks based classifiers is compared with that of the human experts and the Bayes classifier. The ability of the MLP classifier to generalize is far better than that of the Bayes classifier. The efficiency of the neural network based classifiers in conjunction with several types of well-known evoked potential features, such as Fourier transform space, latency and temporal wave, is examined. Although a large clinical data base would be necessary, before this approach can be fully validated, the initial results are promising.

Automatic long-term vigilance analysis systems require information about the occurrence and type of eye movements, in addition to information about other physiological signals. This paper presents a method to detect different types of eye movements in ambulatory recordings. The method is based on the application of a weighted FIR-median-hybrid filter in the preprocessing of the signal and on the novel use of linear correlation between two EOG signals which are obtained using a new, improved electrode montage. The evaluation of the method showed that it performed well in detecting isolated unambiguous eye movements, but differences were observed in comparison to visual scoring in borderline cases. The method was found to be suitable for use as part of a signal analysis system for drowsiness studies.

Auditory brainstem responses are used to detect hearing defects in audiology and otoncurology. The use of computer programs for the analysis of such recordings is increasing. To identify their detailed properties a pattern recognition algorithm implemented in an analysis program must be highly reliable. For the recognition process, some preprocessing phases after recording are necessary, such as filtering and often also segmentation. In the following, we will explore segmentation, which can be used in preprocessing of biomedical signals after filtering. We studied linear segmentation, where slopes of short signal segments are computed and divided into different classes according to their values. A segment length of 8 samples for a sampling frequency of 50 kHz employed was best according to our tests and error criteria. Using clustering, we found that less than 10 segment classes is suitable for pattern recognition.

Manual differentiation of electroencephalography (EEG) paper recordings in cases of depression is not very helpful. So, a Multilayer Perceptron (MLP) has been used to differentiate the EEG power density spectra (qEEG) in the wakeful state from animals (control, exercised and depressed). The qEEG ranging from 1 to 30 Hz, at 1 Hz increments (30 input features) and also as slow, medium and fast activity (represented by three ranges of frequencies at the input) were used. After training with depressed and control qEEG only, the MLP has been found to distinguish successfully between the normal and the depressed rats in more than 80% of the cases, identifying, in the process, most of the exercised groups' EEG as normal. The reduction in the dimension of input features from 30 individual frequencies to 3 frequency bands has produced similar results. The rules generated for making such distinctions have been found to be similar to the clinical views.

A survey was conducted in the year 1994 among medical school hospitals in Japan to know whether there are some written rules or regulations about the use of medical information, and whether they are well matched both to secure patient's privacy and to promote the adequate use of the information. A questionnaire was mailed to 80 medical school hospitals in Japan and 65 of them responded. Besides copies of rules, questioners were also requested for detailed investigation. Twenty nine hospitals responded to our request. The security implementation issues in Japan will be discussed using data thus obtained.

Using two examples of installed applications that are widely spread in a large teaching hospital the awareness of secure communication is highlighted. Teaching to the rotating medical staff is organised on a regular basis. The physicians learn the responsibility they accept when entering the hospital information system (HIS). In a distributed environment, the confidentiality aspects change with the technology when the users perform with on-line helps and graphical interfaces.