Part 1 — Special Issue On: “Frontiers in Development of Intelligent Systems for Human Biomechanics using Soft Computing”No Access

PERFORMANCE EVALUATION OF DRY EYE DETECTION SYSTEM USING HIGHER-ORDER SPECTRA FEATURES FOR DIFFERENT NOISE LEVELS IN IR THERMAL IMAGES

VIDYA K. SUDARSHAN

Department of Biomedical Engineering, School of Science and Technology, Singapore University of Social Science, Singapore City 599491, Singapore

School of Electrical and Computer Engineering, University of Newcastle, Singapore City 038986, Singapore

Search for more papers by this author

JOEL E. W. KOH

Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore City 599489, Singapore

Search for more papers by this author

JEN HONG TAN

Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore City 599489, Singapore

Search for more papers by this author

YUKI HAGIWARA

Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore City 599489, Singapore

E-mail Address: [email protected]

Corresponding author.

Search for more papers by this author

KUANG CHUA CHUA

Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore City 599489, Singapore

Search for more papers by this author

EDDIE Y. K. NG

School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore City 639798, Singapore

Search for more papers by this author

, and

LOUIS TONG

Singapore Eye Research Institute, Singapore City 168751, Singapore

Singapore National Eye Center, Singapore City 168751, Singapore

Duke-NUS Graduate Medical School, Singapore City 169857, Singapore

Yong Loo Lin School of Medicine, National University of Singapore, Singapore City 117597, Singapore

Search for more papers by this author

https://doi.org/10.1142/S0219519417400103Cited by:6 (Source: Crossref)

Abstract

The enhanced tear film evaporation and diminished tear production causes a dry eye (DE) condition. A non-invasive infrared (IR) thermography is most commonly used as a diagnostic tool for diagnosis of DE. However, the availability of high-quality IR thermal camera at low cost is difficult. Hence, an efficient DE detection system which can perform efficiently by using low-cost and low-quality images instead of conventional IR images would be a significant contribution. Therefore, in this work, we have evaluated the performance of automated non-invasive DE detection system using low-quality images obtained by adding different levels of noise to high-quality IR images. In this work, the performances of two non-linear higher-order spectra (HOS) cumulants and bispectrum features are compared. These features are extracted from the IR images with different levels of Gaussian noise. Principal component analysis (PCA) is performed on these extracted features and they are ranked using $t$ -value and later fed to different classifiers. We have achieved the accuracies, sensitivities and specificities of: (i) 86.90%, 85.71% and 88.10%, with noise level 0 using 24 bispectrum features, and (ii) 80.95%, 85.71% and 76.19%, with noise level 10 using 15 bispectrum features for right eye IR images. This study exhibits that even in the presence of high levels of noise, the detection of DE is possible and our proposed method performs efficiently using HOS bispectrum features. Thus, our proposed method can be used to detect DE using low-quality and inexpensive cameras instead of high-cost IR camera.

Keywords:

References

1. Ding J, Sullivan A, Aging and dry eye disease, Exp Gerontol 47 :483–490, 2012. Crossref, Web of Science, Google Scholar
2. Reddy P, Grad O, Rajagopalan K, The economic burden of dry eye: A conceptual framework and preliminary assessment, Cornea 23 :751–761, 2004. Crossref, Web of Science, Google Scholar
3. Miljanovic B, Dana R, Sullivan DA, Schaumberg DA, Impact of dry eye syndrome on vision-related quality of life, Am J Ophthalmol 143 :409–415, 2007. Crossref, Web of Science, Google Scholar
4. Yu J, Asche CV, Fairchild CJ, The economic burden of dry eye disease in the United States: A decision tree analysis, Cornea 30 :379–387, 2011. Crossref, Web of Science, Google Scholar
5. Phadatare SP, Momin M, Nighojkar P, Askarkar S, Singh KK, A comprehensive review of dry eye disease: Diagnosis, medical management, recent developments, and future challenges, Adv Pharm 2015 :704946, 2015. Google Scholar
6. Savini G, Prabhawasat P, Kojima T, Grueterich M, Espana E, Goto E, The challenges of dry eye diagnosis, Clin Ophthalmol 2 :31–55, 2008. Crossref, Google Scholar
7. Bron AJ, Tomlinson A, Foulks GN, Pepose JS, Baudouin C, Geerling G, Nichols KK, Lemp MA, Rethinking dry eye disease: A perspective on clinical implications, Ocul Surf 12 :S1–S31, 2014. Crossref, Web of Science, Google Scholar
8. Rosenfield M, Computer vision syndrome: A review of ocular causes and potential treatments, Ophthalmic Physiol Opt 31 :502–515, 2011. Crossref, Web of Science, Google Scholar
9. Ankita SB, Samir GB, Sunita MJ, A review on recent advances in dry eye: Pathogenesis and management, Oman J Ophthalmol 4 :50–56, 2011. Crossref, Google Scholar
10. Santosh K, Alan T, Angus M, Charles D, Kannu R, Dry eye diagnosis, Invest Ophthalmol Vis Sci 49 :1407–1414, 2008. Crossref, Web of Science, Google Scholar
11. Xu KP, Yagi Y, Toda I, Tsubota K, Tear function index: A new measure of dry eye, Arch Ophthalmol 113 :84–88, 1995. Crossref, Google Scholar
12. Vico E, Benitez CJM, Gimenez RA, Fernandez C, Garcia SJ, Tear function index validation for dry eye diagnosis, Arch Soc Esp Oftalmol 79 :265–271, 2004. Google Scholar
13. Perry HD, Dry eye disease: Pathophysiology, classification, and diagnosis, Am J Manage Care 14 :S79–S87, 2008. Web of Science, Google Scholar
14. Nichols KK, Mitchell GL, Zadnik K, The repeatability of clinical measurements of dry eye, Cornea 23 :272–285, 2004. Crossref, Web of Science, Google Scholar
15. Purslow C, Wolffsogn J, The relation between physical properties of the anterior eye and ocular surface temperature, Optom Vis Sci 84 :197–201, 2007. Crossref, Web of Science, Google Scholar
16. Tan JH, Ng EYK, Acharya UR, Chee C, Infrared thermography on ocular surface temperature: A review, Infrared Phys Technol 52 :97–108, 2009. Crossref, Web of Science, Google Scholar
17. Tan JH, Ng EYK, Acharya UR, Automated study of ocular thermal images: Comprehensive analysis of corneal health with different age group subjects and validation, Digit Signal Process 20 :1579–1591, 2010. Crossref, Web of Science, Google Scholar
18. Tan JH, Ng EYK, Acharya UR, Evaluation of tear evaporation from ocular surface by functional infrared thermography, Med Phys 37 :6022–6034, 2010. Crossref, Web of Science, Google Scholar
19. Tan JH, Ng EYK, Acharya UR, Chee C, Study of normal ocular thermogram using textural parameters, Infrared Phys Technol 53 :120–126, 2010. Crossref, Web of Science, Google Scholar
20. Tan JH, Ng EYK, Ahcarya UR, An efficient automated algorithm to detect ocular surface temperature on sequence of thermograms using snake and target tracing function, J Med Syst 35 :949–958, 2011. Crossref, Web of Science, Google Scholar
21. Morgan PB, Tullo AB, Efron N, Infrared thermography of the tear film in dry eye, Eye 9 :615–618, 1995. Crossref, Web of Science, Google Scholar
22. Fujishima H, Toda I, Yamada M, Sato N, Tsubota K, Corneal temperature in patients with dry eye evaluated by infrared radiation thermometry, Br J Ophthalmol 80 :29–32, 1996. Crossref, Web of Science, Google Scholar
23. Chiang HK, Chen CY, Cheng HY, Chen KH, Chang DO, Development of infrared thermal imager for dry eye diagnosis, Proc SPIE 6294 :629406, 2006. Crossref, Google Scholar
24. Tai YS, Hwa CK, Liu PH, Wu MH, Chang DO, Su PF, Chang SW, Chiang HK, Noncontact detection of dry eye using a custom designed infrared thermal image system, J Biomed Opt 16 :046009, 2011. Crossref, Web of Science, Google Scholar
25. Acharya UR, Tan JH, Koh JEW, Vidya KS, Yeo S, Too CL, Chua KC, Ng EYK, Tong L, Automated diagnosis of dry eye using infrared thermography images, Infrared Phys Technol 71 :263–271, 2015. Crossref, Web of Science, Google Scholar
26. Acharya UR, Tan JH, Vidya KS, Yeo S, Too CL, Lim WJE, Chua KC, Tong L, Diagnosis of response and nonresponse to dry eye treatment using infrared thermography images, Infrared Phys Technol 67 :497–503, 2014. Crossref, Web of Science, Google Scholar
27. Beiger T, Neely S, Feature-based image metamorphosis, Comput Graph 26 :35–42, 1992. Crossref, Google Scholar
28. Gonzalez RC, Woods RE, Digital Image Processing, Prentice Hall, USA, 2002. Google Scholar
29. Acharya UR, Sree SV, Krishnan MMR, Molinari F, Saba L, Ho SYS, Ahuja AT, Ho SC, Nicolaides A, Suri JS, Atherosclerotic risk stratification strategy for carotid arteries using texture-based features, Ultrasound Med Biol 38 :899–915, 2012. Crossref, Web of Science, Google Scholar
30. Helgason S, The Radon Transform, 2nd edn. Birkhauser, Boston, 1999. Crossref, Google Scholar
31. Acharya UR, Faust O, Sree VS, Swapna G, Martis RJ, Kadri NA, Suri JS, Linear and nonlinear analysis of normal and CAD-affected heart rate signals, Comput Methods Prog Biomed 113 :55–68, 2014. Crossref, Web of Science, Google Scholar
32. Nikias CI, Petropulu AP, Higher-Order Spectra Analysis: A Nonlinear Signal Processing Framework, Prentice Hall Signal Processing Series, PTP Prentice Hall, NJ, 1993. Google Scholar
33. Martis RJ, Acharya UR, Mandana KM, Ray AK, Chakraborty C, Cardiac decision making using higher order spectra, Biomed Signal Process Control 8 :193–203, 2013. Crossref, Web of Science, Google Scholar
34. Chandran V, Elgar S, Shape discrimination using invariants defined from higher-order spectra, Proc IEEE Int Conf Acoustics, Speech and Signal Processing, pp. 3105–3108, 1991. Google Scholar
35. Chandran V, Elgar SL, Position, rotation, and scale invariant recognition of images using higher-order spectra, Proc IEEE Int Conf Acoustic, Speech and Signal Processing, Vol. 5, pp. 213–216, 1992. Google Scholar
36. Acharya UR, Kannathal N, Hua LM, Yi LM, Study of heart rate variability signals at sitting and lying postures, J Bodyw Mov Ther 9 :134–141, 2005. Crossref, Google Scholar
37. Acharya UR, Faust O, Kadri NA, Suri JS, Yu WW, Automated identification of normal and diabetes heart rate signals using nonlinear measures, Comput Biol Med 43 :1523–1529, 2013. Crossref, Web of Science, Google Scholar
38. Mendel JM, Tutorial on higher-order statistics in signal processing and system theory: Theoretical results and some applications, Proc IEEE 79 :278–305, 1991. Crossref, Web of Science, Google Scholar
39. Andrew RW, Statistical Pattern Recognition, 2nd edn., John Wiley & Sons, Chichester, UK, 2002. Google Scholar
40. Duda RO, Hart PE, Stork DG, Pattern Classification, 2nd edn., John Wiley & Sons, New York, 2001. Google Scholar
41. Box FJ, Guinness, Gosset, Fisher, and small samples, Stat Sci 2 :45–52, 1987. Crossref, Google Scholar
42. Duda RO, Peter EH, David GS, Pattern Classification, Wiley-Interscience, New York, 2012. Google Scholar
43. Tong L, Seow G, Tan J, Dry eye, BMJ 345 :e7533, 2012. Crossref, Google Scholar
44. Tong L, Waduthantri S, Wong TY, Saw SM, Wang JJ, Rosman M, Lamoureux E, Impact of symptomatic dry eye on vision-related daily activities: The Singapore Malay eye study, Eye 24 :1486–1491, 2010. Crossref, Web of Science, Google Scholar
45. Lam SM, Tong L, Duan XR, Acharya UR, Tan JH, Petznick A, Wenk MR, Shui GH, Longitudinal changes in the levels of tear fluid lysophospholipids and O-acyl-omega-hydroxy fatty acids brought about by eyelid warming treatment in a cohort of Meibomian gland dysfunction patients, J Lipid Res 55(9) :959–1969, 2014, doi:https://doi.org/10.1194/jlr.P051185. Web of Science, Google Scholar
46. Sim HS, Petznick A, Barbier S, Tan JH, Acharya UR, Yeo S, Tong L, A randomized, controlled treatment trial of eyelid-warming therapies in Meibomian gland dysfunction, Ophthalmol Ther 3 :37–48, 2014. Crossref, Google Scholar
47. Acharya UR, Mookiah MRK, Koh JEW, Tan JH, Bhandary SV, Rao AK, Hagiwara Y, Chua KC, Laude A, Automated diabetic macular edema (DME) grading system using DWT, DCT features and maculopathy index, Comput Biol Med 84 :59–68, 2017. Crossref, Web of Science, Google Scholar
48. Acharya RU, Faust O, Alvin APC, Sree SV, Molinari F, Saba L, Nicolaides A, Suri JS, Symptomatic vs. asymptomatic plaque classification in carotid ultrasound, J Med Syst 36 :1861–1871, 2012. Crossref, Web of Science, Google Scholar
49. Acharya UR, Mookiah MRK, Koh JEW, Tan JH, Noronha K, Bhandary SV, Rao AK, Hagiwara Y, Chua KC, Laude A, Novel risk index for the identification of age-related macular degeneration using radon transform and DWT features, Comput Biol Med 73 :131–140, 2016. Crossref, Web of Science, Google Scholar
50. Acharya UR, Mookiah MRK, Koh JEW, Tan JH, Bhandary SV, Rao AK, Fujita H, Hagiwara Y, Chua KC, Laude A, Automated screening system for retinal health using bi-dimensional empirical mode decomposition and integrated index, Comput Biol Med 75 :54–62, 2016. Crossref, Web of Science, Google Scholar
51. Takashi K, Reiko I, Murat D, Eiki G, Yoji T, Yukhihiro M, Minako K, Yuichi O, Kazuo T, A new noninvasive tear stability analysis system for the assessment of dry eye, Invest Ophthalmol Vis Sci 45 :1369–1374, 2004. Crossref, Web of Science, Google Scholar
52. Dorota HS, David AC, Iskander DR, Read SA, Collins MJ, Predicting dry eye using noninvasive techniques of tear film surface assessment, Invest Ophthalmol Vis Sci 52 :751–756, 2011. Crossref, Web of Science, Google Scholar
53. Acharya UR, Faust O, Molinari F, Sree SV, Junnarkar SP, Sudarshan V, Ultrasound-based tissue characterization and classification of fatty liver disease: A screening and diagnostic paradigm, Knowl-Based Syst 75 :66–77, 2015. Crossref, Web of Science, Google Scholar
54. Sree SV, Ng EYK, Acharya RU, Faust O, Breast imaging: A survey, World 2 :171–178, 2011. Google Scholar
55. Acharya UR, Fujita H, Oh SL, Hagiwara Y, Tan JH, Muhammad A, Automated detection of arrhythmias using different intervals of tachycardia ECG segments with convolutional neural network, Inf Sci 405 :81–90, 2017. Crossref, Web of Science, Google Scholar
56. Acharya UR, Fujita H, Oh SL, Muhammad A, Tan JH, Chua KC, Automated detection of coronary artery disease using different durations of ECG segments with convolutional neural network, Knowl-Based Syst (2017), doi:https://doi.org/https://doi.org/10.1016/j.knosys.2017.06.003. Crossref, Web of Science, Google Scholar
57. Acharya UR, Fujita H, Sudarshan VK, Sree VS, Eugene LWJ, Ghista DN, An integrated index for detection of sudden cardiac death using discrete wavelet transform and nonlinear features, Knowl-Based Syst 83 :149–158, 2015. Crossref, Web of Science, Google Scholar
58. Acharya UR, Vidya KS, Ghista DN, Lim WJE, Molinari F, Sankaranarayanan M, Computer-aided diagnosis of diabetic subjects by heart rate variability signals using discrete wavelet transform method, Knowl-Based Syst 81 :56–64, 2015. Crossref, Web of Science, Google Scholar
59. Acharya UR, Faust O, Sree SV, Molinari F, Garberoglio R, Suri JS, Cost-effective and non-invasive automated benign & malignant thyroid lesion classification in 3D contrast-enhanced ultrasound using combination of wavelets and textures: A class of ThyroScan $^{^{TM}}$ algorithms, Technol Cancer Res Treat 10 :371–380, 2011. Crossref, Web of Science, Google Scholar
60. Acharya UR, Sudarshan VK, Adeli H, Santhosh J, Koh JEW, A novel depression diagnosis index using nonlinear features in EEG signals, Eur Neurol 74 :79–83, 2015. Crossref, Web of Science, Google Scholar
61. Acharya UR, Chua EC, Faust O, Lim TC, Lim LF, Automated detection of sleep apnea from electrocardiogram signals using nonlinear parameters, Physiol Meas 32 :287–303, 2011. Crossref, Web of Science, Google Scholar
62. Faust O, Acharya UR, Krishnan SM, Lim CM, Analysis of cardiac signals using spatial filling index and time-frequency domain, Biomed Eng Online 3 :30, 2004. Crossref, Web of Science, Google Scholar
63. Acharya UR, Bhat S, Koh JEW, Bhandary SV, Adeli H, A novel algorithm to detect glaucoma risk index using texton and local configuration pattern features extracted from fundus images, Comput Biol Med 88 :72–83, 2017. Crossref, Web of Science, Google Scholar
64. Acharya UR, Raghavendra U, Fujita H, Hagiwara Y, Koh JEW, Tan JH, Sudarshan VK, Vijayananthan A, Yeong CH, Gudigar A, Ng KH, Automated characterization of fatty liver disease and cirrhosis using curvelet transform and entropy features extracted from ultrasound images, Comput Biol Med 79:250–258, 2016. Crossref, Web of Science, Google Scholar