Wyniki wyszukiwania - BazTech

1

Enabling Autonomous Medical Image Data Annotation: A human-in-the-loop Reinforcement Learning Approach

da Cruz Leonardo, Sierra-Franco César, Silva-Calpa Greis, Raposo Alberto

Annals of Computer Science and Information Systems

|

2021

|

Vol. 25

271--279

EN

Deep learning techniques have shown significant contributions to several fields, including medical image analysis. For supervised learning tasks, the performance of these techniques depends on a large amount of training data as well as labeled data. However, labeling is an expensive and time-consuming process. With this limitation, we introduce a new approach based on Deep Reinforcement Learning (DRL) to cost-effective annotation in a set of medical data. Our approach consists of a virtual agent to automatically label training data, and a human-in-the-loop to assist in the training of the agent. We implemented the Deep Q-Network algorithm to create the virtual agent and adopted the method mentioned above, which employs human advice to the virtual agent. Our approach was evaluated on a set of medical X-ray data in different use cases, where the agent was required to create new annotations in the form of bounding boxes from unlabeled data. Results show that an agent training with advice positively impacts obtaining new annotations from a data set with scarce labels. This result opens up new possibilities for advancing the study and implementing autonomous approaches with human advice to create a cost-effective annotation in data sets for computer-aided medical image analysis.

2

Generating Fuzzy Linguistic Summaries for Menstrual Cycles

Sosnowski Łukasz, Penza Tomasz

Annals of Computer Science and Information Systems

|

2020

|

Vol. 21

119--128

EN

This paper presents a method of generating linguistic summaries of women's menstrual cycles based on the set of concepts describing various aspects of the cycles. These concepts enable description of menstrual cycles that are readable for humans, but they also provide high-level information that can be used as control input for other data processing actions such as e.g. anomaly detection. The labels signifying these concepts are assigned to cycles by means of multivariate time series analysis. The corresponding algorithm is a subsystem of a bigger solution created as a part of an R&D project.

3

Automatic Generation of Annotated Corpora of Diagnoses with ICD-10 codes based on Open Data and Linked Open Data

Boytcheva Svetla, Velichkov Boris, Velchev Gerasim, Koychev Ivan

Annals of Computer Science and Information Systems

|

2020

|

Vol. 21

163--167

EN

We propose methods for automatic generation of corpora that contains descriptions of diagnoses in Bulgarian and their associated codes in ICD-10-CM (International Classification of Diseases, 10th revision, Clinical Modification). The proposed approach is based on the available open data and Linked Open Data and can be easily adapted for other languages. The resulted corpora generated for the Bulgarian clinical texts consists of about 370,000 pairs of diagnoses and corresponding ICD-10 codes and is beyond the usual size that can be generated manually, moreover it was created from scratch and for a relatively short time. Further updates of the corpora are also possible whenever new open resources are available or the current ones are updated.

4

Predicting blood glucose using an LSTM neural network

El Idrissi Touria, Idri Ali, Abnane Ibtissam, Bakkoury Zohra

Annals of Computer Science and Information Systems

|

2019

|

Vol. 18

35--41

EN

Diabetes self-management relies on the blood glucose prediction as it allows taking suitable actions to prevent low or high blood glucose level. In this paper, we propose a deep learning neural network model for blood glucose prediction. The model is a sequential one using a Long- Short-Term Memory (LSTM) layer with two fully connected layers. Several experiments were carried out over data of 10 diabetic patients to decide on the model's parameters in order to identify the best variant of the model. The performance of the proposed model measured in terms of root mean square error (RMSE) was compared with the ones of an existing LSTM model and an autoregressive (AR) model. The results show that our model is significantly more accurate; in fact, our LSTM model outperforms the existing LSTM model for all patients and outperforms the AR model in 9 over 10 patients, besides, the performance differences were assessed by thWilcoxon statistical test. Furthermore, the mean of the RMSE of our model was 12.38 mg/dl while it was 28.84 mg/dl and 50.69 mg/dl for AR and the existing LSTM respectively.

5

MOPS - a feasibility study for working with GPS and sensor data in a medical context

Meigen Christof, Vogel Mandy, Bumberger Jan

Annals of Computer Science and Information Systems

|

2019

|

Vol. 18

621--624

EN

New kinds of data collection like GPS-tracking, wearable sensors and mobile apps impose both technical and privacy challanges for medical research. In the MOPS study (''Machbarkeitsstudie für Ortsbezogene Parameter und Sensordaten'' - feasibility study for geocoded parameters and sensor data) we provide participants with a newly developed app and sensors for various physical and environmental parameters. We want to explore the feasibility of the recently established Medical Research Platform of the Medical Faculty of the University of Leipzig and similar platforms for this kind of data collection and processing.

6

Monitoring of the natural plague foci using immunocomputing

Sokolova S. P., Abdullina V. Z., Sokolova L. A.

Systems Science

|

2005

|

Vol. 31, no 2

107-115

EN

This paper discusses the development of an Immunocomputing (IC) approach to infection control of natural plague foci. The following concepts are developed: the structures and mathematical concepts of temporal relational databases; the concept of applying a plague risk index (PRI); using data fusion and mathematical models to form a PRI using an Immunocomputing approach. The application of Immunocomputing indices can reduce large quantities of variable data relating to a complex interacting dynamic system, into a single general value or index that represents all of those factors (data fusion) to obtain a solution to a practical problem.