Hyperspectral imaging fluorescence excitation scanning for colon cancer detection

Leavesley, Silas J.; Walters, Mikayla G.; López, Carmen; Baker, Thomas; Favreau, Peter F.; Rich, Thomas C.; Rider, Paul; Boudreaux, Carole

doi:10.1117/1.jbo.21.10.104003

Cited by 58 publications

(38 citation statements)

References 77 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As suggested above, an SRDA could therefore be integrated as a "chip-on-tip" at the distal end of a commercial endoscope in future. This is beneficial when compared to multispectral systems based on filter wheels 21,30,31 and tunable filters, 23,32 which sacrifice temporal resolution to sequentially acquire spectral data. Sequential acquisition of spectral data can also introduce motion artifacts between sequentially acquired spectral band images.…”

Section: Discussionmentioning

confidence: 99%

“…Methods with a finer sampling of the spectral response, such as multi-or hyperspectral imaging (MSI/HSI) systems (≈10 and 100 spectral color channels, respectively), 12,13 may enable quantitative assessment of functional tissue properties and more detailed statistical classification of the spectral changes that occur during disease. The use of reflectance-based MSI/HSI systems in biomedical applications has, for example, been shown to increase contrast in vascular conditions, 14 wound healing, [15][16][17] ophthalmology, 15,18,19 cancer diagnostics, 12,[20][21][22][23] and for the determination of tumor resection margins. 24 The application of a targeted fluorescent contrast agent is often referred to as optical molecular imaging (OMI) and has shown promise for endoscopic cancer detection in the GI tract.…”

Section: Introductionmentioning

confidence: 99%

“…Spectral imaging is typically achieved by filtering the light on either the illumination or detection side. Filtering has been achieved by the use of: a set of dichroic beamsplitters and dedicated detectors; 10,27,29 a filter wheel with a set of bandpass filters; 21,30,31 tunable filters; 23,32 or a monochromator. 33 Alternatively, a set of laser lines may be used for illumination.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Bimodal reflectance and fluorescence multispectral endoscopy based on spectrally resolving detector arrays

et al. 2018

View full text Add to dashboard Cite

Emerging clinical interest in combining standard white light endoscopy with targeted near-infrared (NIR) fluorescent contrast agents for improved early cancer detection has created demand for multimodal imaging endoscopes. We used two spectrally resolving detector arrays (SRDAs) to realize a bimodal endoscope capable of simultaneous reflectance-based imaging in the visible spectral region and multiplexed fluorescencebased imaging in the NIR. The visible SRDA was composed of 16 spectral bands, with peak wavelengths in the range of 463 to 648 nm and full-width at half-maximum (FWHM) between 9 and 26 nm. The NIR SRDA was composed of 25 spectral bands, with peak wavelengths in the range 659 to 891 nm and FWHM 7 to 15 nm. The spectral endoscope design was based on a "babyscope" model using a commercially available imaging fiber bundle. We developed a spectral transmission model to select optical components and provide reference endmembers for linear spectral unmixing of the recorded image data. The technical characterization of the spectral endoscope is presented, including evaluation of the angular field-of-view, barrel distortion, spatial resolution and spectral fidelity, which showed encouraging performance. An agarose phantom containing oxygenated and deoxygenated blood with three fluorescent dyes was then imaged. After spectral unmixing, the different chemical components of the phantom could be successfully identified via majority decision with high signal-to-background ratio (>3). Imaging performance was further assessed in an ex vivo porcine esophagus model. Our preliminary imaging results demonstrate the capability to simultaneously resolve multiple biological components using a compact spectral endoscopy system.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bimodal reflectance and fluorescence multispectral endoscopy based on spectrally resolving detector arrays

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Various HSI technical approaches have been developed for biomedical applications-most based on designs that generate reflectance and/or fluorescence maps over a continuous spectral bandwidth. [1][2][3] Hyperspectral content can be used in conjunction with computational models of light transport and various statistical methods, such as principal component analysis, to calculate optical and physiological properties for each pixel. [4][5][6] However, these methods generally require assumptions or approximations about certain tissue features, such as scattering properties and water concentration.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral imaging in the spatial frequency domain with a supercontinuum source

et al. 2019

View full text Add to dashboard Cite

imaging in the spatial frequency domain with a supercontinuum source," J.Abstract. We introduce a method for quantitative hyperspectral optical imaging in the spatial frequency domain (hs-SFDI) to image tissue absorption (μ a ) and reduced scattering (μ 0 s ) parameters over a broad spectral range. The hs-SFDI utilizes principles of spatial scanning of the spectrally dispersed output of a supercontinuum laser that is sinusoidally projected onto the tissue using a digital micromirror device. A scientific complementary metaloxide-semiconductor camera is used for capturing images that are demodulated and analyzed using SFDI computational models. The hs-SFDI performance is validated using tissue-simulating phantoms over a range of μ a and μ 0 s values. Quantitative hs-SFDI images are obtained from an ex-vivo beef sample to spatially resolve concentrations of oxy-, deoxy-, and met-hemoglobin, as well as water and fat fractions. Our results demonstrate that the hs-SFDI can quantitatively image tissue optical properties with 1000 spectral bins in the 580-to 950-nm range over a wide, scalable field of view. With an average accuracy of 6.7% and 12.3% in μ a and μ 0 s , respectively, compared to conventional methods, hs-SFDI offers a promising approach for quantitative hyperspectral tissue optical imaging.

show abstract

“…focus on the visible parts of the spectrum, especially focusing on the blue and green wavelengths. It has been shown in Leavesley et al[17] that for colon cancer, the blue-violet portion of the spectrum (390 nm -450 nm) has the highest correlation with instances of cancer.…”

mentioning

confidence: 98%

Classification of Hyperspectral Colon Cancer Images Using Convolutional Neural Networks

Mobilia

View full text Add to dashboard Cite

CLASSIFICATION OF HYPERSPECTRAL COLON CANCER IMAGES USING CONVOLUTIONAL NEURAL NETWORKS by Sean J. Mobilia Hyperspectral images are 3-D images, which contain data in hundreds of spectral bands as opposed to 2-D images, which contain data in at most 3 bands (red, green, and blue). Hyperspectral imagery was initially developed for remote sensing; however, recently, researchers have started to see its potential in medical diagnosis and cancer detection. Hyperspectral images provide massive amounts of data about the objects they are studying, and this causes challenges during information processing. Machine learning tools, such as convolutional neural networks (CNNs), are known to be successful in extracting features and classifying traditional 2-D images. This thesis proposes CNN architectures for processing hyperspectral data for colon cancer detection. Using data taken from a limited number of colon tissue samples, this thesis shows that the proposed CNN architecture can classify cancerous and noncancerous tissue samples utilizing hyperspectral information. The obtained results are compared to grayscale images of the same tissue samples, looking both at grayscales of the individual hyperspectral bands and panchromatic grayscale images in which the spectral bands are merged together. The CNN using the hyperspectral data shows advantages over the grayscale data, with a 5.6% improvement in accuracy and a 0.037 improvement in F1 score over the individual band grayscale images and a 21.7% improvement in accuracy and a 0.178 improvement in F1 score over the panchromatic grayscale images. The results are also compared to a K-nearest neighbor (KNN) classifier and a logistic regression (LR) classifier using the hyperspectral data, and the CNN shows advantages over both. The CNN has a 17.9% improvement in accuracy and a 0.141 improvement in F1 score over the KNN classifier and a 5% improvement in accuracy and a 0.061 improvement in F1 score over the LR classifier. ACKNOWLEDGMENTS I would like to thank my advisor, Professor Birsen Sirkeci, for all her help and support during this project and for having tremendous patience with me. I would also like to thank my other committee members, Professor Robert Morelos-Zaragoza and Professor Chang Choo, for taking the time to read and review my work.

show abstract

Hyperspectral imaging fluorescence excitation scanning for colon cancer detection

Cited by 58 publications

References 77 publications

Bimodal reflectance and fluorescence multispectral endoscopy based on spectrally resolving detector arrays

Bimodal reflectance and fluorescence multispectral endoscopy based on spectrally resolving detector arrays

Hyperspectral imaging in the spatial frequency domain with a supercontinuum source

Classification of Hyperspectral Colon Cancer Images Using Convolutional Neural Networks

Contact Info

Product

Resources

About