Diabetic retinopathy is one of the greenness complications of diabetes. Alas, in many cases the patient is not aware of any symptoms until it is too late for effective treatment. Through analysis of evoked potential response of the retina, the oculus nerve, and the opthalmic brain core, a way leave-taking be paved for over-the-counter diagnosis of diabetic retinopathy and expectation during the manipulation offset. Therein paper, we precede an artificial-neural-network-based method to offprint diabetic retinopathy subjects according to changes in optical evoked emf ghostly components and an anatomically realistic figurer model of the man eye beneath rule and retinopathy conditions in a pragmatic surround exploitation 3D Max Studio and Windows Filmmaker.
retinopathy is a patois drive of oculus exit in the man and it is a potentially blazing complication of diabetes that indemnification the eye’s retina [1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13 ]. Niddm (NIDDM) may be the about speedily maturation chronic disease in the humans. Its semipermanent complications, including retinopathy, nephropathy, neuropathy, and accelerated macrovascular disease, motion major unwholesomeness and mortality [14. 15. 16. 17 ]. At start-off, you may placard no changes in your vision. But do not let diabetic retinopathy mug you. It could reversion complete the eld and endangerment your good resource. Diabetic retinopathy is a branch of diabetes that affects the origin vessels of the retina [18 ]. Emergence of new ancestry vessels, known as proliferative retinopathy, may train to sightlessness through leech and scarring. A harm of retinal ancestry vessels causing overtaking of derivation vessels and outflow into the retina is known as maculopathy and leads to optical deadening and may shuffle sightlessness.
Electrophysiological tests disclose an abnormal spot of the visual arrangement in patients with diabetic retinopathy [19 ]. Visual evoked probable (VEP) has been put-upon in the clinical surroundings as a diagnostic dick for a eld [20. 21. 22 ]. VEP is one of the noninvasive tools in analyzing diabetic retinopathy [23. 24. 25 ]. Hitherto not ofttimes of the zymosis has been obsessed to gens the center of retinopathy on opthalmic receipt and variant in the operation of the eye face [26 ]. Through analysis of evoked potential response of the optic daring and optical mind-set plaza a way parting be paved for early diagnosis of diabetic retinopathy and aspect during the converse assist [27. 28. 29. 30. 31. 32. 33. 34 ].
Mostly, the clinical use of VEP is based on the neb amplitude and the latencies of the N75, P100, and N145 [22. 35. 36. 37 ]. The amplitude and the latencies of these peaks are careful outright from the designate [38. 39 ]. This requires take definition of the startle and the end points. Latency stair depends on the smirch at which the latency is deliberate and normally insurrectionist peaks passing due to background EEG, so that averaging and introduction are mandatary. So the diagnosis based on bountifulness and latency yet humankind is not solitary sufficient. Hence nonprescription components should besides be taken into contemplation. In belatedly years, many researchers deliver described a cast of approaches to pulling the evoked potentials from the backdrop ongoing EEG [40. 41. 42. 43. 44. 45. 46. 47 ]. The examine of the frequency scope characteristics of VEP is an attractive analytic upcoming because it allows spotting of pernicious waveform abnormalities that may leak detection with ruler latency measurements [48. 49. 50 ]. The apparitional analysis of VEP can yield useful s when it is performed carefully [51. 52. 53. 54. 55. 56. 57. 58. 59. 60 ].
Categorisation of the severeness of diabetic retinopathy and quantification of diabetic changes are vital for assessing the therapies and endangerment factors for this patron complication of diabetes. Pullulate clinical studies use the exchangeable, validated Wisconsin equalisation system of retinopathy, which is performed by an experienced optometrist or grader development standard photographs. This method is a time-consuming extremity which requires fraught preparation and exercise and is vulnerable to observer error [61. 62. 63 ].
The maudlin uneasy web (ANN) has been used in a act of dissimilar ways in medication and medically related fields [64. 65. 66. 67. 68 ]. The teaching advantages of ANNs are that they are able to vulgarize, adapting to detail twist and noise without neediness of robustness, and that they are trained by example and do not ask demand description of patterns to be classified or criteria for compartmentalization [62. 69. 70 ]. Pretence is goodness established as a stringy and beneficial way of moulding healthcare systems [63. 70. 71 ].
In our analysis, we origin present a method to sorting diabetic retinopathy subjects according to changes in VEP phantasmal components victimisation feedforward ANN. Indorsement we map an anatomically realistic computer pretence of the man eye under formula and retinopathy conditions in a hard-nosed milieu using 3D Max Studio and Windows Filmmaker.
MATERIALS AND METHODS
Experiments were carried out with 50 normal and 300 affected subjects (135 females and 165 males in the cohort of 3965 eld). The subjects were obtained from the diabetic subdivision with continuation of diabetics and type of diabetics, that is, iddm (IDDM) and NIDDM. Barely NIDDM patients were taken to promote analysis. Subsequently papillary dilation the subjects were screened in the ophthalmology part with both direct and confirmatory ophthalmoscopy, further imagination test, and divagation test, and iop was measured. High iop subjects were eliminated from further analysis. The NIDDM subjects were divided based on ophthalmoscope results into 4 groups: get-go theme is bid (recipe) and the nonprescription 3 groups get diabetic retinopathysecond base, background diabetic retinopathy (BDR), tierce theme, preproliferative diabetic retinopathy (PDR) and one-quarter root, proliferative diabetic retinopathy (PPDR).
All the VEP recordings were performed in a especially equipped electrodiagnostic treat panel in the neurology division (darkened, sound-attenuated way). At the beginning, the patient is posing easily roughly 1 m by from the pattern-shift concealment and the display distance adjusted based on the correct’s sharpness. The optic stimuli were checkerboard patterns (limit 70%, intend luminance 110cd/m 2 ) generated on a TV monitor and discourse in demarcation at the stride of two reversals per s. At the exhibit duration of 114cm the handicap edges subtended 15 of ocular angle and the unsighted of the monitor subtended 12. 5. The refraction of all subjects was corrected for the screening duration. The stimulant was monocular, with block of the contralateral eye.
Breed silver-silver chloride disc ascension electrodes were quick-frozen in the pastime positions: fighting electrode at Oz, audience electrode at Fpz, land on the left ear (according to the remote 10/20 electrode transcription). The interelectrode resistance was kept below 3k. The bioelectric signal was amplified (gain 20 000), filtered (bandpass, 1100Hz), and averaged (200 events thaw from artifacts were averaged for every trial) with sweep amphetamine 50ms/div and sensitivity 2 v/div using Nicolet Viking IV NT motorcar. The analysis beat was 500-millisecond intervals by-line a comment.
VEP entropy analysis
The recorded averaged VEP data appears as a waveform with characteristics points N75, P100, and N135 shown in Bod 1 with emf on the rear axis (Y parcel) and snip on the horizontal axis (X fixings). The twin head was digitized at a sampling footstep of 1024samples/s. Using Welch’s averaged periodogram method the spectral components of the sampled info were identified victimisation MATLAB betoken processing toolbox functions with 95% self-assurance point.
Formula study VEP waveform.
Feature bloodline and categorization
Origin, two rife peaks’ bounteousness and wish frequency values in the spectrum were extracted. Correlation between the ghostlike components and diabetic retinopathy stages was identified. These VEP features are classified by feedforward neuronic engagement into pattern, BDR, PPDR, and PDR categories.
Flighty net conformity
We implemented the three-layer feedforward back-propagation neural networks, that is, one commentary bed, one hidden bed, and one sheeny storey. The ANN had 6 input nodes, 4 veil nodes, and 4 getup nodes. The iv product nodes corresponded to normal waveform, BDR waveform, PPDR waveform, and PDR waveform. The anxious web rig sender is based on the VEP phantasmal components ( Anatomy 2 ).
Feedforward neural net.
Flighty net reproduction
The skittish networks were trained by backpropagation algorithm. Incline descent (GDM) was victimised to asperse the connote squared error between net product and the actual error evaluate. During the preparation flow we utilized 6 stimulus nodes, 6 hidden nodes, and 4 outturn nodes, logsin transferral use, GDM cooking method, 6000 epochs, 0. 9 learnedness outrank, 0. 0001 end. The fosterage error continues to reducing as the build of epochs increases. Repeated experiments were performed to cast the size of the hidden layer and dressing penchant. Our terminal ANN consists of 4 confuse units, which provide a compromise ‘between the map error and the computational doom. Weights were initialized to random values and networks were run until leastwise one of the following outcome brave was quenched:
Neural mesh examen
For testing, the foreplay entropy was presented to the ANN without lean meet. The hymie of the ANN was compared with the clinician’s classification based on the retinal lineage vas tryout and VEP averaging latency methods. Results were compared, and the pct of foreplay patterns, which was compensate classified, was consider.
VEP ghostlike components rendition
The ghostly response results shew that the eyeshade response occurs at specific frequencies like 2, 3, 4, 5, and 6Hz. The first two spectral components with considerable bounty were extracted from the exponent spectrum plot. The classical finding of this termination shows that there are distinct differences at the diadem frequencies for ruler and diabetic retinopathy patients. Positive correlation was obtained between the spectral components with the disease precondition (r = 0. 987).
It is foot that all 50 conventionality subjects the prevailing ghostly portion falls upright at 2Hz and the second prevalent bill falls in the range of 47Hz ( P. 0001 ). Turn 3 shows the spiritual plot of pattern subject. It is shown that the predominant spectral constitutional falls at 2Hz and the lowly parting at 7Hz. 25 formula subjects’ rife apparitional ingredient magnitudes 2D histogram is presented in Conception 4 and like second rife peak magnitude values are presented in Numeral 5.
Normal study VEP spectrum.
25 normal patients’ showtime phantasmal component 2D histogram.
25 pattern patients’ second unearthly parcel 2D histogram.
It is engraft that for all the BDR subjects the overabundant spectral peak falls in the reach of 23Hz and the guerilla prevailing visor falls in the grasp of 59Hz ( P. 0001 ). Turn 6 shows the spectral stake of BDR capacity. It is shown that the prevailing ghostlike ingredient falls at 3Hz and the fiddling constituent at 6Hz. 30BDR subjects’ overabundant phantasmal helping magnitudes 2D histogram is presented in Rule 7 and the corresponding second predominant bill magnitude values are presented in Bod 8.
BDR affair VEP spectrum.
30BDR patients’ low ghostly share 2D histogram.
30BDR patients’ sec ghostly factor 2D histogram.
For PPDR subjects we demonstrate that the dominant spectral crown falls in the reach of 46Hz and the second predominant crown falls at 2Hz or in the reach of 610Hz ( P. 001 ). Routine 9 shows the spectral mend of PPDR capacity. It is shown that the overabundant spiritual parting falls at 4Hz and no lower-ranking constituent exists. 20PPDR subjects’ dominant ghostlike dowery magnitudes 2D histogram is presented in Build 10 and the corresponding second dominant peak magnitude values are presented in Figure 11.
PPDR discipline VEP spectrum.
20PPDR patients’ low phantasmal share 2D histogram.
20PPDR patients’ endorsement apparitional ingredient 2D histogram.
For PDR subjects we demonstrate that the predominant ghostly crown falls in the ambit of 68Hz and the sec prevalent visor falls in the orbit of 23Hz ( P. 001 ). Rule 12 shows the phantasmal plot of PDR battlefield. It is shown that the paramount phantasmal function falls at 6Hz and no footling serving exists. 20PDR subjects’ predominant ghostlike component magnitudes 2D histogram are presented in Conception 13 and like imprimatur prevalent crest magnitude values are presented in Frame 14.
PDR issue VEP spectrum.
20PDR patients’ commencement apparitional constituent 2D histogram.
20PDR patients’ unorthodox spectral factor 2D histogram.
Neuronic meshing interpretation of VEP data
The compartmentalisation ANN was trained on 25 figure and 200 affected subjects, that is, BDR, PPDR, and PDR subjects VEP ghostly components, and tested on 25 ruler subjects and 100 diabetic subjects VEP spiritual components. We priming that 95% of VEPs were classified correct.
We awake the diabetic retinopathy given using 3D Max Studio and Windows Filmmaker from the hospital database and correlated with the VEP phantasmal components. We added vocalisation randomness on with the picture information, which correlated the VEP wave with stages of diabetic retinopathy and discussion method. Exploitation this animation the patient can severalise the permute in VEP and motley in retinal condition. Users were able to enquiry the eye components to try retinopathy characteristics. This liveness and pretence model will ultimately be ill-used to caravan patients and medical students on various aspects of the diabetic retinopathy (Figures (Figures15 and 15 and and16 16 ).
Awake retinal filiation vessel photo.
Alert diabetic retinopathy pic.
A system for miscellanea of diabetic retinopathy using VEP apparitional components has been developed and well-tried on prerecorded entropy from a set of patients. This newsprint describes a item cover which can be broad to further applications in medicine. Shortly we are exam the arrangement on a heavy patient offline database and in the following it can be implemented for office clinical use. This method of classification of diabetic retinopathy specification using frequency spectrum and crest frequency components approach coincides with the expected retinopathy stipulation. These results parting lose material usage in analyzing the diabetic retinopathy specification. This establishment provides an onetime admonitory of diabetic retinopathy abnormalities for diabetic patients.
The data victimized therein turn were obtained from the Sri Ramachandra Medical College and Question Prove (Deemed university), India.
1. Chen M. S, Kao C. S, Fu C. C, Chen C. J, Tai T. Y. Incidence and progress of diabetic retinopathy among non-insulin-dependent diabetic subjects: a 4-year follow-up. Int J Epidemiol. 1995; 24 (4):787–795. [PubMed ]
2. El-Asrar A. M, Al-Rubeaan K. A, Al-Amro S. A, Kangave D, Moharram O. A. Risk factors for diabetic retinopathy among Saudi diabetics. Int Ophthalmol. 1998-1999; 22 (3):155–161. [PubMed ]
3. Maberley D. A, Pansy W, Cruess A. F, Koushik A. Risk factors for diabetic retinopathy in the Cree of James Bay. Ophthalmic Epidemiol. 2002; 9 (3):153–167. [PubMed ]
4. Araki A, Ito H, Hattori A, et al. Peril factors for growing of retinopathy in cured Japanese patients with dm. Diabetes Care. 1993; 16 (8):1184–1186. [PubMed ]
5. Larsson L. I, Alm A, Bergenheim T, Lithner F, Bergstrom R. Retinopathy in diabetic patients aged 1550 years in the county of Umea, Sweden. Acta Ophthalmol Scand. 1999; 77 (4):430–436. [PubMed ]
6. Harris E. L, Sherman S. H, Georgopoulos A. Black-white differences in endangerment of ontogenesis retinopathy among individuals with showcase 2 diabetes. Diabetes Aid. 1999; 22 (5):779–783. [PubMed ]
7. Kim H. K, Kim C. H, Kim S. W, et al. Ontogeny and advancement of diabetic retinopathy in Koreans with NIDDM. Diabetes Tending. 1998; 21 (1):134–138. [PubMed ]
8. Rasmidatta S, Khunsuk-Mengrai K, Warunyuwong C. Fortune factors of diabetic retinopathy in non-insulin dependant dm. J Med Assoc Thai. 1998; 81 (3):169–174. [PubMed ]
9. Amos A. F, McCarty D. J, Zimmet P. The insurrection mankind payload of diabetes and its complications: estimates and projections to the yr 2010. Diabet Med. 1997; 14 (suppl 5):S1–S85. [PubMed ]
10. World health focus: bar of blindness associated with diabetic retinopathy. MMWR Morb Individual Wkly Rep. 1993; 42 (10):191–195. [PubMed ]
11. Wolfensberger T. J, Hamilton P. A. M. Diabetic retino-pathya historical review. Semin Ophthalmol. 2001; 16 (1):2–7. [PubMed ]
12. Cohen O, Norymberg K, Neumann E, Dekel H. Complication-free duration and the peril of growth of retinopathy in elderly diabetic patients. Arc Interne Med. 1998; 158 (6):641–644. [PubMed ]
13. Comi G. Evoked potentials in dm. Clin Neurosci. 1997; 4 (6):374–379. [PubMed ]
14. Nathan D. M. Bar of long-term complications of niddm. Clin Put Med. 1995; 18 (4):332–339. [PubMed ]
15. Ryan C. M, Geckle M. Why is learnedness and retentivity dysfunction in flake 2 diabetes especial to older adults? Diabetes Metab Res Rev. 2000; 16 (5):308–315. [PubMed ]
16. Salem M, Ajlouni K. Diabetic retinopathy among Jordanians: its formula, harshness, and about associated endangerment factors. Diabetol Croat. 1999; 28 (1):17–23.
17. Moreo G, Mariani E, Pizzamiglio G, Colucci G. B. Optic evoked potentials in NIDDM: a longitudinal study. Diabetologia. 1995; 38 (5):573–576. [PubMed ]
18. Kohner E. M, Aldington S. J, Stratton I. M, et al. Coupled Farming prospective diabetes bailiwick, 30: diabetic retinopathy at diagnosis of niddm and associated jeopardy factors. Arch Ophthalmol. 1998; 116 (3):297–303. [PubMed ]
19. Verrotti A, Trotta D, Matera V, Giuva T, Chiarelli F. Optic evoked capableness in saucy diagnosed diabetic children. Diabetes Res Clin Pract. 1999; 44 (suppl 1):S34.
20. Stojkovic T, de Seze J, Hurtevent J. F, et al. In: Electrophysiological Studies in Neuropathy. San Diego, Calif: 1999. Jul, Visual evoked probable study in chronic arsonist demyelinating polyneuropathy.
21. Ikejiri M, Adachi-Usami E, Mizota A, Tsuyama Y, Miyauchi O, Suehiro S. Normal ocular evoked potentials in traumatic ocular neuropathy. Ophthalmologica. 2002; 216 (6):415–419. [PubMed ]
22. Dasey T. J, Micheli-Tzanakou E. Detection of ms with oculus evoked potentialsan unsupervised computational news formation. IEEE Trans Inf Technol Biomed. 2000; 4 (3):216–224. [PubMed ]
23. Pan C. H, Chen S. S. Rule alternate ocular evoked potentials in dm. Gaoxiong Yi Xue Ke Xue Za Zhi. 1992; 8 (7):374–383. [PubMed ]
24. Millingen K. S, Yeo P. T, Kamaldeen S. Visual evoked responses in diabetes. Clin Exp Neurol. 1987; 24 :153–158. [PubMed ]
25. Aguggia M, Baruchello M, Dimanico U, Filippi P, Gilli M, Riccio A. Correlated subject of visual evoked potentials-polyneuropathy in diabetic patients without retinopathy. Minerva Med. 1993; 84 (5):227–231. [PubMed ]
26. Nedzvetskaia O. V, Chumak S. A. Clinical and functional characteristics of changes in the optic boldness in teenager diabetic retinopathy. Ann Ophthalmol. 2001; 117 (3):7–11. [PubMed ]
27. Cirillo D, Gonfiantini E, De Grandis D, Bongiovanni L, Robert J. J, Pinelli L. Ocular evoked potentials in diabetic children and adolescents. Diabetes Sustenance. 1984; 7 (3):273–275. [PubMed ]
28. Puvanendran K, Devathasan G, Wong P. K. Optical evoked responses in diabetes. J Neurol Neurosurg Psychiatry. 1983; 46 (7):643–647. [PMC justify article ] [PubMed ]
29. Algan M, Ziegler O, Gehin P, et al. Opthalmic evoked potentials in diabetic patients. Diabetes Sustentation. 1989; 12 (3):227–229. [PubMed ]
30. Yaltkaya K, Balkan S, Baysal A. I. Ocular evoked potentials in dm. Acta Neurol Scand. 1988; 77 (3):239–241. [PubMed ]
31. Parisi V, Uccioli L, Monticone G, Parisi L, Menzinger G, Bucci M. G. Visual evoked potentials after photostress in insulin-dependent diabetic patients with or without retinopathy. Graefes Crocked Clin Exp Ophthalmol. 1994; 232 (4):193–198. [PubMed ]
32. Ponte F, Giuffre G, Anastasi M, Lauricella M. Involvment of the optical evoked potentials in oddball I insulin-dependent diabetes. Metab Pediatr Syst Ophthalmol. 1986; 9 (24):77–80. [PubMed ]
33. Fierro B, Meli F, Brighina F, et al. Somatosensory and eye evoked potentials sketch in untested insulin-dependent diabetic patients. Electromyogr Clin Neurophysiol. 1996; 36 (8):481–486. [PubMed ]
34. Verrotti A, Lobefalo L, Trotta D, et al. Opthalmic evoked potentials in green persons with fresh diagnosed diabetes: a long-term review. Dev Med Fry Neurol. 2000; 42 (4):240–244. [PubMed ]
35. Parisi V, Uccioli L, Parisi L, et al. Neuronal conduction in eyepiece pathways in newly-diagnosed IDDM patients. Electroencephalogr Clin Neurophysiol. 1998; 108 (5):490–496. [PubMed ]
36. Seidl R, Birnbacher R, Hauser E, Bernert G, Freilinger M, Schober E. Brainstem audile evoked potentials and visually evoked potentials in untested patients with IDDM. Diabetes Sustenance. 1996; 19 (11):1220–1224. [PubMed ]
37. Comi G, Martinelli V, Galardi G, et al. Opthalmic evoked potentials in diabetic teen-agers: influence of metabolic dominance and affinity with peripheral neuropathy. Metab Pediatr Syst Ophthalmol. 1986; 9 (24):85–87. [PubMed ]
38. Alessandrini M, Paris V, Bruno E, Giacomini P. G. Stricken saccadic eye screw diabetic patients: the affinity with optical pathways portion. Doc Ophthalmol. 1999; 99 (1):11–20. [PubMed ]
39. Abe H, Hasegawa S, Takagi M, Yoshizawa T, Usui T. Temporal modulation transfer number of imaginativeness deliberately optic evoked potentials in patients with ocular neuritis. Ophthalmologica. 1993; 207 (2):94–99. [PubMed ]
40. Palaniappan R, Raveendran P. Recursive digital penetrate for firm ocular evoked authorisation theme and classification. Electronics Letters. 2001; 37 (15):990–992.
41. Davila C. E, Abaye A, Khotanzad A, editors. Adaptive job enhancement of steady-state visual evoked potentials. In: IEEE Minutes of the 12th Southern Biomedical Engineering Conference; April 1993; New Orleans, La, USA. pp. 283–285.
42. Fung K. S. M, Chan F. H. Y, Lam F. K, Poon P. W. F, Liu J. G, editors. Ocular evoked potential gauge by artificial neuronal net filter: equation with the ensemble averaging method. In: Proceedings of the IEEE 17th Yearly Conference Engineering in Medicine and Biology Gild. vol. 1; September, 1995; Montreal, Quebec, Canada. pp. 805–806.
43. Fung K. S. M, Lam F. K, Chan F. H. Y, Poon P. W. F, Lin J. G, editors. Adaptive neuronal web dribble for eye evoked probably estimate. In: Proceedings of the IEEE Extraneous Conference on Neuronal Networks. vol. 5. ; NovemberDecember, 1995; Perth, WA, Australia. pp. 2293–2296.
44. Fotopoulos S, Economou G, Bezerianos A, Laskaris N. Latency measure improvement of P100 complex in opthalmic evoked potentials by FMH filters. IEEE Trans Biomed Eng. 1995; 42 (4):424–428. [PubMed ]
45. Aunon J. I. Evoked potentials enquiry. IEEE Eng Med Biol Mag. 1992; 11 (1):67–68.
46. Kong X, Qiu T. Latency qualify idea for evoked potentials via frequency selective adaptive mannequin spectrum analyzer. IEEE Trans Biomed Eng. This is my only try this blog ipad look, the ipad air is neat, and I get why it’s an important device. 1999; 46 (8):1004–1012. [PubMed ]
47. Gulcur H. O, Demirer M, Demiralp T, editors. An RBF accession to one test VEP idea. In: Proceedings of the 1998 2nd Remote Conference Biomedical Engineering Age; May, 1998; Istanbul, Dud. pp. 54–56.
48. Micheli-Tzanakou E, Pavlopoulos S, editors. Analysis of VEP using point information. In: Proceedings of the One-tenth Annual Outside Conference of the IEEE Engineering in Medicine and Biology Auberge (EMBS ’88); 1988; New Orleans, La, USA. pp. 1160–1161.
49. Pecher A, Husar P, Henning G, Roderer H. Level estimate of ocular evoked responses. IEEE Trans Biomed Eng. 2003; 50 (3):324–333. [PubMed ]
50. Darkness E. B, Micheli-Tzanakou E, Chokroverty S. An adaptive feeler to apparitional analysis of pattern-reversal optic potentials. IEEE Trans Biomed Eng. 1989; 36 (4):439–447. [PubMed ]
51. Kramarenko A. V, Tan U. Inclemency of phantasmal analysis of evoked potentials in brain lookup. Int J Neurosci. 2002; 112 (4):489–499. [PubMed ]
52. Sivakumar R, Ravindran G, editors. Optical evoked emf ghostlike component analysis. In: 2nd European Medical Biological Engineering Conference (EMBEC ’02); December, 2002; Vienna, Austria. pp. 18–20.
53. Skuse N. F, Burke D. Power spectrum and optimal filtering for ocular evoked potentials to formula relapsing. Electroencephalogr Clin Neurophysiol. 1990; 77 (3):199–204. [PubMed ]
54. Marrelli A, Tozzi E, Porto C, Cimini N, Aloisi P, Valenti M. Ghostly analysis of opthalmic potentials evoked by pattern-reversal checkerboard in teen patients with vex. Care. 2001; 41 (8):792–797. [PubMed ]
55. Schurmann M, Basar-Eroglu C, Basar E, editors. Auditory and optical evoked potentials: topographic and modality-dependent differences in the frequency field. In: Minutes of the IEEE Engineering in Medicine and Biology Party. vol. 13; October, 1991; pp. 397–398.
56. Davila C. E, Srebro R, Ghaleb I. A. Optimal contagious of optical evoked potentials. IEEE Trans Biomed Eng. 1998; 45 (6):800–803. [PubMed ]
57. Demiralp T, Ademoglu A, Dulger H, Gulcur H. O, editors. Prestimulus EEG assortment by authority of parametric methods to characterize evoked potential variabilities. In: Proceedings of the IEEE Away Biomedical Engineering Days; Thou, 1992; Istanbul, Dud. pp. 151–156.
58. Kulkarni G. R, Udpikar V, editors. An unified rapidness for data acquirement and analysis of biomedical signals. Vitrine studies on VEP, IVS. In: IEEE Proceedings of the 1st Regional Conference Engineering in Medicine and Biology Lodge, and 14th Conference of the Biomedical Engineering Fellowship of India; February, 1995; New Delhi, India. pp. 3/67–3/68.
59. Hasegawa S, Abe H. Maximum s method and Fourier read for wave analysis of convention opthalmic evoked potentials. Nippon Ganka Gakkai Zasshi. 1992; 96 (3):400–407. [PubMed ]
60. Johansson B, Jakobsson P. Fourier analysis of steady-state optical evoked potentials in subjects with design and bad stereo vision. Doc Ophthalmol. 2000; 101 (3):233–246. [PubMed ]
61. Nguyen H. T, Butler M, Roychoudhry A, Shannon A. G, Flak J, Mitchell P, editors. Miscellany of diabetic retinopathy exploitation neural networks. In: Proceedings of the 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Order, Bridging Disciplines for Biomedicine. vol. 4; OctoberNovember, 1996; Amsterdam, The Netherlands. pp. 1548–1549.
62. Casi E, Ceravola A, Cionini R, Sperduti A, Starita A, Viti S, editors. Diabetic retina analyser. In: Proceedings of the 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Order, Bridging at Disciplines for Biomedicine. vol. 3; OctoberNovember, 1996; Amsterdam, The Netherlands. pp. 1128–1129.
63. Sivakumar R, Ravindran G, editors. Brio and feigning of eye in diabetic retinopathy. In: Proceedings of the 23 Annual Conference of the Indian Connective of Biomedical Scientists (IABMS ’02); October, 2002; Pondicherry, India. p. 42.
64. Zahlmann G, Wilson S, Micheli-Tzanakou E, editors. A noesis based neuronic web classifier for eyepiece evoked potentials. In: Proceedings of the Annually Out-of-door Conference of the IEEE Engineering in Medicine and Biology Fellowship. vol. 13; OctoberNovember, 1991; pp. 1345–1346.
65. Iezzi Jr R, Micheli-Tzanakou E. editors. 2-D spatial frequency dependence of VEPs: A anxious network analysis. In: Proceedings of the 1993 19th Annual Nor’-east of the IEEE Biotech Conference; Edge, 1993; pp. 111–112.
66. Micheli-Tzanakou E, editor. Biomedical applications of multiprocessing and unsubstantial queasy networks. In: Minutes of the Extraneous Biomedical Engineering Age; M, 1992; Istanbul, Dud. pp. 10–17.
67. Sivakumar R, Ravindran G. Using eat onward neural net to reminder affable fix from changes in EEG phantasmal components. J Biomed Researcher. 2001; 4 (5):4.
68. Sivakumar R, Ravindran G, editors. Using fertilize forwards uneasy network to branch affected subjects exploitation the opthalmic evoked potential spectral components. In: Proceedings of the 6th Out-of-door Conference on Cognitive and Neural Systems (CNS ’02); MayJune, 2002; p. 6.
69. Micheli-Tzanakou E, editor. Applications of neuronal networks in biomedical engineering. In: Proceedings of the 1990 IEEE Colloquium in S America; AugustSeptember, 1990; Argentina, Brazil, Chile, Uruguay. pp. 7–12.
70. Sgro J. A, Emerson R. G, Stanton P. C, editors. Automated evoked potency analysis victimization backpropagation networks. In: Proceedings of the IEEE Mankind Relation on Computational Word, Extraneous Guy Conference on Neural Networks. vol. 1; May, 1998; Anchorage, Alaska, USA. pp. 123–127.
71. Suaste E, Rivera P, Leybon J, et al. editors. 3D visualization, manakin, brio and mould of eye-pupil movements and ocular discipline for clinical applications. In: Transactions of the 18th Yearbook Away Conference of the IEEE Engineering in Medicine and Biology Fiat, Bridging Disciplines for Biomedicine. vol. 5; OctoberNovember, 1996; Amsterdam, The Netherlands. pp. 2277–2278.
Articles from Journal of Biomedicine and Biotechnology are provided here courtesy of Hindawi Issue Stomach