Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

The 5th IEEE International Conference on E-Health and Bioengineering - EHB 2015

Grigore T. Popa University of Medicine and Pharmacy, Iaşi, Romania, November 19-21, 2015

An Approach to the Verilog-based System for


Medical Image Enhancement
Iuliana CHIUCHISAN1
Affiliation 1: Department of Computer Science, Automation and Electronics, “Stefan cel Mare” University Suceava, USV,
Suceava, Romania, iuliana.chiuchisan@usm.ro

Abstract—This paper presents a direct implementation and technology is digital due to the use of smart sensors,
improvement of the real-time configurable system for image computer technology, advanced medical equipment, portable
enhancement using Verilog Hardware Description Language medical devices connected in an Internet of Things (IoT)
and reconfigurable architecture (field programmable gate network [4]. In most medical applications, images are
array). New series of filters are developed at the hardware level
acquired directly from the patient and often these images are
for image processing (edges detection, sharpen operation,
enhance contrast operation and brightness-adjustment), in order distorted by noise or other defects which may occur during
to improve the quality of images and to assist in diagnosis the acquisition [4]. The image enhancement techniques and noise
medical specialists. For verification and simulation of the suppression are useful, especially in imaging pre-processing
Verilog-based system for image enhancement was used ISIM level. The image acquisition techniques used in the medical
Simulator, a component of the ISE Design Suite program, from field are: ultrasound (ultrasound and ultrasound tomography -
Xilinx. Using Hardware Description Languages for image the physical principle of reflection at tissues interfaces), X-
processing is a quite new approach extending the field of digital rays (radiology, X tomography - the principle of radiation
design on reconfigurable circuits to digital image processing absorption or attenuation depending on the tissues density),
using VLSI technologies. Describing the image enhancement
chemical factors (specific tracers, radioactive
techniques using Verilog HDL enables rapid prototyping of
these complex algorithms offering the direct possibility of FPGA pharmaceuticals) that are mounted to the investigated tissue
implementation. and emit ionizing radiation (scintigraphy, Single Photon
Emission Computed Tomography - SPECT) electromagnetic
Keywords—medical image enhancement, Verilog hardware field (MRI CT – the principle of nuclear magnetic resonance
description language, real-time configurable system, field of protons in a magnetic field) [4]. These technologies can be
programmable gate array. used to assess the current condition of the patient’s organ or
tissue and can monitor the patient over time for diagnosis and
treatment evaluation [5].
I. INTRODUCTION Medical equipment perform processing of acquired images
Digital image processing (DIP) is used in various fields of through filtering or removing background noise or artefacts,
modern society and especially in medical imaging and, correction of inhomogeneity, distortion of images, numerical
recently, occupies a fundamental role in assisted diagnosis of evaluation of features, visualization of areas of interest etc.
diseases, developing as a multidisciplinary field [1]. DIP is a Image processing software and hardware applications can
very broad field based on rigorous mathematical theory [2] provide temporal and spatial analysis and help to
and implementations of DIP on different computing machines automatically identify and analyze images in order to detect
are resources consuming [3], especially if we consider the patterns or characteristics indicative of tumors. The image
real-time processing of the information extracted from processing and analysis can be used in medical applications
images. to determine, for example, “the diameter, volume and
Nowadays, the image processing is an emerging biomedical vasculature of a tumor or organ, flow parameters of blood or
tool contributing to the improvement of healthcare sector. other fluids and microscopic changes” [4].
Recent image processing tools can perform image analysis, Image processing is “one of the best suited applications for
image enhancement, noise reduction, image segmentation, FPGAs, due to the parallel architectures, the logic density and
geometric transformations, and image restoration. the generous memory capacity” [6]. “The FPGAs are capable
Biomedical imaging focuses on the capture and display of to boost the performance of all the fast imaging algorithms:
the medical images for diagnostic, rehabilitation and image construction, enhancement and restoration, image
treatment purposes. The development of more efficient analysis and pattern recognition, image and data compression,
equipment for medical imaging is closely linked to the color space conversion, etc”. Implementing these applications
research results of the processing and analysis of medical on reconfigurable hardware offers much greater speed than a
images, or more generally multidimensional signals with software implementation [7]. Hardware implementation has
applications in medical imaging [4]. Modern imaging become an attractive alternative for image processing,

978-1-4673-7545-0/15/$31.00 ©2015 IEEE


especially for real time processing due to the advances in very In this paper we added new filters for image enhancement
large scale integration (VLSI) technologies [8]. and we tested the hardware architecture of the real-time
As shown in Figure 1, “a typical diagnostic imaging system configurable system for digital image enhancement. In order
consists of three sets of cards: data acquisition, data to achieve the simulation using ISIM Simulator component
consolidation, and image/data processing cards” [9]. that requires a digital image and involves vectors in
numerical form to be applied to the Verilog-based system, an
application in C # was developed (Fig. 2).

Figure 1 Altera solution for diagnostic imaging equipment [9]

II. PRIOR RESEARCH


In previous research, references [10], [11], [12], a medical Figure 2 DIP Verilog for image enhancement visualization
image processing system was proposed using Verilog HDL
and FPGA (Field Programmable Gate Array) based the novel This application allows loading the image in “bitmap” or
hardware architecture of a real-time configurable system for “jpeg” format and generates a file which contains the digital
digital image processing. This system was used to obtain real- image (RGB vectors for each pixel).
time results more suitable than the original images with We described a new series of image processing: enhance
applications in medical devices. contrast operation, brightness-adjustment, sharpen operation,
In reference [13], a health care system for monitoring of edges detection and included them into real-time configurable
patients at risk in a smart environment was proposed, using system for image processing. Filtering principles of image
an ICT-based solution and Internet of Things (IoT) enhancement techniques were analyzed mathematically and
technologies. In article [14] an approach to developing a then transcribed into a source code written in Verilog
decision support and home monitoring system for patients Hardware Description Language. The filters were described
with neurological disorders using Internet of Things concepts at Register Transfer Level (RTL) using Verilog continuous
was presented. assignments. The images were processed by switching the
In the article [15] was presented a health care system for order of filters, using a single command and without the
Neurological Disorders Screening and Rehabilitation. The reconfiguration of the FPGA, in order to improve the original
system manages data acquired from the patients with image.
Parkinson’s disease in order to support physicians in
always @(posedge clk)begin
diagnosis, treatment and monitoring the patients and also to
facilitate the interaction at the distance between doctors and RGB_Output[7:0]=
filter_orderSharpen==3?RoutSharpen:
patients. filter_orderContrast==3?RoutContrast:
Other proposed system [16] combines data about patient’s filter_orderEdges==3?RoutEdges:RoutBrightness;
medical history, anthropometrical measurements acquired RGB_Output[15:8]=
using the KinectTM sensor for measuring abdominal filter_orderSharpen==3?GoutSharpen:
circumference, biochemical data or other medical filter_orderContrast==3?GoutContrast:
filter_orderEdges==3?GoutEdges:GoutBrightness;
examinations, in order to describe an algorithm capable to
elaborate and generate a personalized diet adapted to one type RGB_Output[23:16]=
filter_orderSharpen==3?BoutSharpen:
of cancer. filter_orderContrast==3?BoutContrast:
III. VERILOG-BASED SYSTEM FOR IMAGE ENHANCEMENT filter_orderEdges==3?BoutEdges:BoutBrightness;
end
The main purpose of this paper is to improve and simulate
the hardware architecture of the real-time configurable The real-time configurable system for image enhancement
system for digital image processing using Verilog HDL and described in Verilog HDL was tested using test-bench module
reconfigurable architecture (FPGA). and synthesized on a FPGA-based platform. The synthesis on
The configurable system for real-time image processing the platform Xilinx Virtex-6 FPGA ML605 of a dental
was described in [10] and [11]. Some point operations for radiography image, using the Verilog-based system
medical image enhancement were described and tested in described, has generated an improved image with a delay of
reference [12]. 18,87 ns.
IV. EXPERIMENTAL RESULTS
Various image processing techniques, such as threshold The simulation was carried out under different conditions:
operation, intensity level slicing, contrast-adjustment, image- changing filter parameters and changing the order of the
negative, Fourier-transform and segmentations, can be used filters, without reconfiguration of the FPGA:
for analysis of medical image in order to detect tumors or  enhance contrast operation, brightness-adjustment,
other structures. sharpen operation and edges detection (Fig. 4, a);
The real-time configurable system architecture described in  sharpen operation, edges detection, enhance contrast
Verilog HDL was illustrated in Figure 3(a). This Verilog- operation and brightness-adjustment (Fig. 4, b);
based system for image enhancement was simulated and  enhance contrast operation, edges detection, sharpen
verified using ISIM Simulator, a component of ISE Design operation and brightness-adjustment (Fig. 4, c);
Suite from Xilinx [17] (Fig. 3, b)  sharpen operation, brightness-adjustment, enhance
The Verilog-based system for image enhancement was contrast operation and edges detection (Fig. 4, d).
tested on a medical image (dental radiography with 860x740
pixels, 8-bit image, size 621Kb [18]), shown in Figure 2.

a)

b)
Figure 3 (a) General architecture of the real-time configurable system; (b) Simulation results for brightness-adjustment
using ISIM Simulator

a) b)
c) d)
Figure 4 (a) ÷ (d) Results of image processing using the Verilog-based system with image enhancement filters: edges detection;
sharpen operation; enhance contrast operation and brightness-adjustment

[8] V.R. Daggu, P. Shruti, A.B. Naveen, V. Muthukumar,


V. CONCLUSION “Implementation and Evaluation of Image Processing Algoritms on
Reconfigurable Architecture using C-based Hardware Descriptive
This paper presents the improvement of the Verilog-based Languages,” International Journal of Theoretical and Applied
system for image processing previously described. New series Computer Sciences, 1(1), pp. 9-34, 2006.
of filters were developed at the hardware level for image [9] FPGA CPLD and ASIC from Altera
https://www.altera.com/solutions/industry/medical/overview.html
processing (edges detection, sharpen operation, enhance [10] I. Chiuchisan, “A New FPGA-based Real-Time Configurable System
contrast operation and brightness-adjustment), in order to for Medical Image Processing”, Proceedings of IEEE International
improve the quality of images and to assist in diagnosis the Conference on e-Health and Bioengineering (EHB2013), ISBN 978-1-
medical specialists. The advantage of this approach is that the 4799-2373-1, 2013.
[11] I. Chiuchisan, O. Geman, “A review of HDL-based system for real-
implementation at the hardware level of our system can lead time image processing used in tumors screening,” Proceeding of the
to a superior image processing speed and do not require a 18th International Conference on System Theory, Control and
dedicated signal processor (DSP) in order to process the Computing (ICSTCC2014), ISBN: 978-1-4799-4602-0, pp. 217-220,
images. For modelling, optimization and simulation of image 2014.
[12] I. Chiuchisan, “Implementation of Medical Image Processing
processing in Verilog HDL was used ISE Design Suite Algorithm on Reconfigurable Hardware,” Proceedings of IEEE
program, from Xilinx. International Conference on e-Health and Bioengineering
The greater potential of this system lies in fact that it is not (EHB2013), ISBN 978-1-4799-2373-1, 2013.
necessary to use an additional processor dedicated to image [13] I. Chiuchisan, H.N. Costin, O. Geman, “Adopting the Internet of
Things Technologies in Health Care Systems”, Proceedings of
processing, which could slow down the flow: image International Conference and Exposition on Electrical and Power
acquisition → preprocessing the image → image use in order Engineering (EPE2014), Workshop on Electromagnetic Compatibility
to establish a decision which can be human-type (in the case and Engineering in Medicine and Biology, pp. 532-535, 2014.
of a diagnosis based on medical images) or non-human (in the [14] I. Chiuchisan, O. Geman, “An Approach of a Decision Support and
Home Monitoring System for Patients with Neurological Disorders
case of an industrial process). using Internet of Things Concepts”, WSEAS Transactions on Systems,
REFERENCES Issue: Multi-models for Complex Technological Systems, 13, pp. 460-
469, 2014.
[1] J.C. Russ, The Image Processing Handbook, 6th Edition, CRC Press, [15] O. Geman, S. Sanei, I. Chiuchisan, A. Prochazka, O. Vysata,
2011. “Towards an Inclusive Parkinson’s Screening System,” Proceedings of
[2] W. Burger, M.J. Burge, Principles of Digital Image Processing, International Conference on System Theory, Control and Computing
Fundamental Techniques, Springer Ed., 2009. (ICSTCC2014), pp. 470-475, 2014.
[3] D.G. Bailey, Design for Embedded Image Processing on FPGAs, John [16] O. Geman, I. Chiuchisan, A.C. Iuresi, Iulian Chiuchisan, M. Dimian,
Wiley & Sons Ed., 2011. A. Bosancu, D. Buliga, M. Covasa, “Intelligent System for a
[4] H.N. Costin, “Recent trends in Medical Image Processing,” Computer Personalized Diet of Obese Patients with Cancer”, Proceedings of
Science Journal of Moldova, 22, no.2 (65), 2014. International Conference and Exposition on Electrical and Power
[5] H.N. Costin, C. Rotariu, I. Alexa, et al., “TELEMON – A Complex Engineering (EPE2014), Workshop on Electromagnetic Compatibility
System for Real Time Medical Telemonitoring,” Proceedings of the and Engineering in Medicine and Biology, pp. 528-531, 2014.
World Congress on Medical Physics and Biomedical Engineering, [17] Xilinx Inc., www.xilinx.com.
Munich, Germany, 25, pp. 92-95, 2009. [18] Medical Images Database: http://imagej.nih.gov/ij/images/mri-
[6] Y. Said, T. Saidani, M. Atri, “FPGA-based Architectures for Image stack.zip.
Processing using High-Level Design”, WSEAS Transactions on Signal
Processing, 11, E-ISSN: 2224-3488, 2015.
[7] Praveen Vanaparthy, G. Sahitya, Krishna Sree, C.D. Naidu, “FPGA
implementation of image enhancement algorithms for biomedical
image processing”, International Journal of Advanced Research in
Electrical, Electronics and Instrumentation Engineering, 2(11), 2013.

You might also like