Papers by Dragan Jankovic
This paper describes the system for collecting data about transport comfort. Those data are recei... more This paper describes the system for collecting data about transport comfort. Those data are received via 9DOF sensor board, which gives Euler angles (yaw, pitch, and roll), as well as the values of 3-axes accelerometer, gyroscope, and magnetometer. These values are intended for use with Android application that collects and displays them, along with the GPS coordinates. Also, the user has the ability to subjectively assess the current comfort level.
Acta Medica Medianae, Dec 15, 2018
This paper presents the analysis of use and acceptance of the medical information system (MIS) wi... more This paper presents the analysis of use and acceptance of the medical information system (MIS) within the primary healthcare. Analysis is mostly based on data from the Health Care Center Niš, although the conclusion was made on data from a dozen health care centers which use the medical information system MEDIS.NET developed within the Laboratory for Medical Informatics at the Faculty of Electronic Engineering in Niš. Analysis of the use of MIS is based on calculating the percentage of successfully entered records of visits, provided medical services, recipes, referral letters and physical examinations. In the context of the analysis of the use of MIS, successfully entered medical service is actually the service that was not changed or deleted later. Results of this analysis are significant for further technical development of the medical information system, and support the identification of these functionalities that are hardly accepted by the end-users and should be further developed. The acceptance of MIS is analyzed in the light of the technology acceptance model. Registration of provided services and keeping the record of physical examinations are taken as representative functionalities. Registration of provided services has been observed as a functionality that is accepted by the users due to simplicity of use (perceived ease of use), while the registration of physical examinations is observed as functionality is presumed to be accepted by the users as useful (perceived usefulness). For the functionalities with the expected acceptance based on the simplicity of use, the rate of correct data input is over 90% in each of the category. However, the rate of correct data inputs for visits and provided services is more than 99%. This is very significant having in mind the fact that these functionalities are often used and the high rate of incorrect inputs would slow down the work of doctors. On the other hand, the percentage of use of special functionalities for input of physical examinations varies considerably. Specially designed functionality for the most common physical examinations of children is used in more than two thirds of cases (sometimes more than 97%), while for the registration of adult's physical examinations the percentage is lower than 20%. Since the users could input data on physical examinations using the form for visit input, as well as with special form, they will probably use the other option only when the frequency of use is high enough or when the improved functionality of the specialized form provides improved system performances. Under users of MIS we consider the medical staff which uses MIS functionalities in accordance with their duties and privileges (doctors, nurses, medical technicians, etc.).
Computer Science and Information Systems
A selection of Computer Science, Informatics or similar study programs for academic studies evide... more A selection of Computer Science, Informatics or similar study programs for academic studies evidently becomes an emerging choice of a vast number of students in recent years. To address some of the open questions, we performed an empirical study based on a survey, with a goal to find out the main motivating factors directing students to select computer science, informatics or similar programs for studying in a much greater extent. The survey was conducted on a sample of 1517 students from five well established, and most traditional faculties of computer science and informatics at three largest university cities in Serbia: Belgrade, Novi Sad, and Nis. The created sample is representative enough to illustrate the current situation and trends common for many similar societies. Our first analysis shows that the main motivating factor to select computer science or informatics study program at all faculties is the students? motivation to study just that topic, while at management faculty ...
Facta Universitatis, Series: Mathematics and Informatics, 2020
The development of information technology increases its use in various spheres of human activity,... more The development of information technology increases its use in various spheres of human activity, including healthcare. Bundles of data and reports are generated and stored in textual form, such as symptoms, medical history, and doctor’s observations of patients' health. Electronic recording of patient data not only facilitates day-to-day work in hospitals, enables more efficient data management and reduces material costs, but can also be used for further processing and to gain knowledge to improve public health. Publicly available health data would contribute to the development of telemedicine, e-health, epidemic control, and smart healthcare within smart cities. This paper describes the importance of textual data normalization for smart healthcare services. An algorithm for normalizing medical data in Serbian is proposed in order to prepare them for further processing (F1-score=0,816), in this case within the smart health framework. By applying this algorithm, in addition to t...
A bunch of patient data in hospital information systems is kept every day. A part of the data tha... more A bunch of patient data in hospital information systems is kept every day. A part of the data that contains the symptoms, anamnesis, and observations of a doctor can be used for further processing in order to improve public health. This paper describes the algorithm for normalizing medical textual data in order to prepare them for further processing, and within the framework of smart healthcare services. The result of its application is normalized data but also a corpus of stop words specific to the medical domain.
The e-health platform Pinga is an integrated health information system represents a central elect... more The e-health platform Pinga is an integrated health information system represents a central electronic system, in which all medical and health related information about patients, health workers, facilities, documents, and procedures is stored and processed. The platform is implemented in Serbia as MojDoktor (My Doctor), and Macedonia as MojTermin (My Appointment). The architecture of the system was designed to allow for process oriented development with agile methodologies. This methodology allowed for fast deployment and adoption but a change in the architecture to a more formal approach is required to assure its extensibility, soundness, interoperability and standardization. In this paper we propose a transformation of the design framework from Process Oriented to a Model-Driven Architecture.
This paper proposes a secured radio-frequency identification for smart healthcare environment wit... more This paper proposes a secured radio-frequency identification for smart healthcare environment with specialized system functionalities in order to enhance the quality of health service by improving healthcare, unique patient identification, specialized message delivery and activity related with patients rescue. Big Data part is presented for the analytic and data processing purposes in Health Care Information System, for massive data analysis and their visualization. In this paper we demonstrate how proposed prototype of information system can significantly improve daily emergency and healthcare operations and shows its benefits.
Named entity recognition is a widely used task to extract various kinds of information from unstr... more Named entity recognition is a widely used task to extract various kinds of information from unstructured text. Medical records, produced by hospitals every day contain huge amount of data about diseases, medications used in treatment and information about treatment success rate. There are a large number of systems used in information retrieval from medical documentation, but they are mostly used on documents written in English language. This paper contains the explanation of our approach to solving the problem of extracting disease and drug names from medical records written in Serbian language. Our approach uses statistical language models and can detect up to 80% of named entities, which is a good result given the very limited resources for Serbian language, which makes the process of detection much more difficult.
IJEEC - INTERNATIONAL JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTING, 2020
Thanks to medical information systems, many medical reports are collected in an electronic form d... more Thanks to medical information systems, many medical reports are collected in an electronic form daily. Apart from the fields with allowed values for input (the structural part), one part of these reports consists of the free, non-structural text. It contains a more detailed description of the patient's condition, which could not be described using the structural part. Symptoms, results of laboratory analyses, accompanying diagnoses, etc. can often be found in it. Due to a lack of time, doctors often write these descriptions in non-standard ways, using their abbreviations and synonyms, and they often contain typos. All this makes it difficult to extract information in documents specific to the medical domain. This paper presents the creation of medical lexical resources for the automatic labeling of terms from diagnoses in medical reports. In order to perform the automatic marking of the free text, methods of the computer processing of natural languages are needed, as well as app...
Croatian Journal of Education - Hrvatski časopis za odgoj i obrazovanje, 2015
Getting feedback from external stakeholders, especially alumni and employers, is a challenging ta... more Getting feedback from external stakeholders, especially alumni and employers, is a challenging task for higher education institutions. It is important for many reasons, such as providing information to policy makers, assessing quality of academic programmes, adapting academic programmes to labour market needs, and marketing purposes. Alumni and employer surveys are usually used as methods for collecting data from stakeholders. Research suggests that major problems in data collection are related to efficiency and effectiveness of data collection process. This paper presents a different method of data collection supported by ICT. A specially designed web portal equipped with advanced alumni tracking and communication features was developed and used for alumni data collection. The web portal was evaluated after a period of real operation by analyzing collected data and conducting an online survey. The evaluation results showed that the web portal could be used as effective, efficient, ...
In this paper we propose a method for optimization of fixed polarity arithmetic expressions (FPAE... more In this paper we propose a method for optimization of fixed polarity arithmetic expressions (FPAEs) based on dual polarity. The method exploits a simple relationship between two FPAEs for dual polarities. It starts from the zero polarity FPAE of the given function and calculates all FPAEs using dual polarity route. Conversion from one FPAE to another one is carried out using one-bit checking i.e. by using a very simple rule. Therefore, the proposed method is efficient. Experimental results show this.
Facta Universitatis, Series: Automatic Control and Robotics, 2017
This paper is concerned with the complexity of the configuration process in small-and medium-size... more This paper is concerned with the complexity of the configuration process in small-and medium-sized one-of-a-kind production (SMEs OKP) with the dominant variation of the topology from the aspect of the position of customer involvement decoupling point (CIDP), batch size and the complexity of the product configurations process. The issues of the level of the product adjustment to the customer, complexity of the product configuration process, the possibility for the client to be integrated in that early phase of the production and the way of using the data obtained in the product configuration process for generating the final product specification of the customised product in this type of production have been discussed. A special concern has been given to the product complexity paradox, i.e. a risk that a customer may decide not to purchase the product because of too many technical details they encounter while configuring the product, as well as an inevitable extension of the leading ...
arXiv (Cornell University), May 16, 2022
During the process of medical information system development, resource awareness is neglected. It... more During the process of medical information system development, resource awareness is neglected. It is often assumed that the underlying hardware will always have enough memory, processing power, and network bandwidth. Unfortunately, this approach seems not so feasible in every case, and these assumptions, if proven wrong, will harm the initial development run, a later system upgrade, and life cycle in general. This paper aims to raise a resource awareness problem that could still influence all information system deployment and maintenance steps. As an example, we described the influence of the general hardware and network limitations on the information system design, functionality update process, and external system integration. Our research results are a set of guidelines that should be applied to support resource awareness, as an important part of the system design.
Electronics
Complex industrial systems run the different pieces of software in several interconnected physica... more Complex industrial systems run the different pieces of software in several interconnected physical layers. The software update in such an environment must be performed in the shortest possible period with the lowest possible resource usage. Namely, it is critical to minimize the data traffic, decrease software downtime, and reduce the impact of the transitional stage during the update process. To meet these requirements and to unify the update process, the common software node structure along with a hybrid software deployment strategy is proposed. The hybrid strategy is defined as a combination of the existing and well-tested techniques—blue/green, canary, and feature flags. The main aim was to utilize their positive sides and to obtain a better overall effect. The approach was tested in the simulation environment, based on the realistic factory layout, and running the software related to the enterprise resource planning (ERP) level. For successful updates, the proposed hybrid deplo...
In this paper, we present a system based on feature extraction techniques and image segmentation ... more In this paper, we present a system based on feature extraction techniques and image segmentation techniques for detecting and diagnosing abnormal patterns in breast thermograms. The proposed system consists of three major steps: feature extraction, classification into normal and abnormal pattern and segmentation of abnormal pattern. Computed features based on Gray Level Co-occurrence Matrices (GLCM) are used to evaluate the effectiveness of textural information possessed by mass regions. A total of 20 GLCM features are extracted from thermograms. The ability of feature set in differentiating abnormal from normal tissue is investigated using a Support Vector Machine classifier, Naive Bayes classifier and K-Nearest Neighbor classifier. To evaluate the classification performance, five-fold cross validation method and Receiver operating characteristic analysis was performed. The verification results show that the proposed algorithm gives the best classification results using K-Nearest Neighbor classifier and a accuracy of 92.5 %. Image segmentation techniques can play an important role to segment and extract suspected hot regions of interests in the breast infrared images. Three image segmentation techniques: minimum variance quantization, dilation of image and erosion of image are discussed. The hottest regions of thermal breast images are extracted and compared to the original images. According to the results, the proposed method has potential to extract almost exact shape of tumors.
To improve development of the medical information system components, we have introduced model bas... more To improve development of the medical information system components, we have introduced model based code generation tool. It was of great help when we needed to develop series of components sharing the same set of basic functionalities – predominantly data collection forms. Since we faced many functionality update requests, we tend to check the usability of our model driven approach in cases when legacy components developed under the older version of same programming framework need to be included in our system. In this paper we described the set of updates needed for our code generation components as well as two most important use cases – when existing functionality from legacy system has to be extended and when significant portion of legacy system needs to be adapted and included in new system. Overall conclusion is that model driven approach is still useful even with upgrade projects, but required side effort is higher, especially when the first component in a row is getting adapted.
The main intent of our research was to examine the overall acceptance rate among medical professi... more The main intent of our research was to examine the overall acceptance rate among medical professionals by comparing a number of records entered by medical and administrative staff members. To keep results more objective, we analyzed data collected from January 1st, 2012 until December 31st, 2015. Different acceptance rates were observed among different departments. Differences are explained in the scope of the technology acceptance model, based on the different influence of the external properties. In several departments and sub-departments, organizational structure and lack of IT infrastructure make administrative workers the only persons that can use information system. For these departments, a number of records registered by administrative workers can be assumed as potential false positives, thus they are presented separately. Thanks to this research, we are now able to restructure our deployment strategies and to work closely with our potential users to improve healthcare workfl...
Having experience with different traditional and agile methodologies in medical information syste... more Having experience with different traditional and agile methodologies in medical information system development and upgrade projects, we support the statement that no single methodology can be used with its full potential in all cases. Instead, mentioned methodologies have their own place in complex software projects and within certain category of sub-projects give the best results. In this paper we presented the guideline on choosing adequate methodology in different development and upgrade projects. All examined projects are defined within the scope of larger medical information system called Medis.NET developed by our research group, and deployed in 30 healthcare institutions. The bottom-line is that during requirement collection we rely on model and feature driven development supported by automatic code generation tools, which lead to faster prototype generation. Next, dividing the set of functionalities to sub-projects we group them in three categories – core projects, userorien...
During exploitation of a medical information system dedicated to primary care facilities, we real... more During exploitation of a medical information system dedicated to primary care facilities, we realized that significant amount of often used, but rarely changed data is scattered in large number of data tables requiring many join operations when need to be retrieved. Additionally, in many cases, many data fields are retrieved by select queries and later neither displayed on user interface, nor used in any other way. This situation could be a cause for some performance issues as well as unnecessary increase of data traffic. For this reason we decided to improve our system by applying command-query responsibility segregation (CQRS) pattern with de-normalized read database in order to reduce time and data amount needed for some often executed queries. In read database design process we applied model driven approach and used our existing data modeling, mapping and code generating tools. Also, we developed a synchronization component responsible for migrating data from main to read databa...
Both in medical information system development and upgrade project, we often face with a challeng... more Both in medical information system development and upgrade project, we often face with a challenge of creating large number of reports and data collection forms. In order to reduce our efforts in this segment of system development, we tried to use various code generation and reporting tools. The main problem with standardized tools were lack of flexibility. Thus, we decided to develop domain model based framework that consists of data modeling, inverse engineering, code generation and model interpretation libraries and tools. Data modeling tool is used to create domain specific model starting from the loaded meta model. Both code generation and runtime interpretation tools use domain specific model as a basic input, and together with visual templates and generation/interpretation classes form easily extendable and customizable system. In our medical information systems development and upgrade projects we use both approaches and tend to define their proper roles in overall informatio...
Uploads
Papers by Dragan Jankovic