A novel visible and near-infrared hyperspectral imaging platform for automated breast-cancer detection.

BACKGROUND: Breast cancer is a leading cause of cancer-related deaths among women worldwide. Early and accurate detection is crucial for improving patient outcomes. Our study utilizes Visible and Near-Infrared Hyperspectral Imaging (VIS-NIR HSI), a promising non-invasive technique, to detect cancerous regions in ex-vivo breast specimens based on their hyperspectral response. METHODS: In this paper, we present a novel HSI platform integrated with fuzzy c-means clustering for automated breast cancer detection. We acquire hyperspectral data from breast tissue samples, and preprocess it to reduce noise and enhance hyperspectral features. Fuzzy c-means clustering is then applied to segment cancerous regions based on their spectral characteristics. RESULTS: Our approach demonstrates promising results. We evaluated the quality of the clustering using metrics like Silhouette Index (SI), Davies-Bouldin Index (DBI), and Calinski-Harabasz Index (CHI). The clustering metrics results revealed an optimal number of 6 clusters for breast tissue classification, and the SI values ranged from 0.68 to 0.72, indicating well-separated clusters. Moreover, the CHI values showed that the clusters were well-defined, and the DBI values demonstrated low cluster dispersion. Additionally, the sensitivity, specificity, and accuracy of our system were evaluated on a dataset of breast tissue samples. We achieved an average sensitivity of 96.83%, specificity of 93.39%, and accuracy of 95.12%. These results indicate the effectiveness of our HSI-based approach in distinguishing cancerous and non-cancerous regions. CONCLUSIONS: The paper introduces a robust hyperspectral imaging platform coupled with fuzzy c-means clustering for automated breast cancer detection. The clustering metrics results support the reliability of our approach in effectively segmenting breast tissue samples. In addition, the system shows high sensitivity and specificity, making it a valuable tool for early-stage breast cancer diagnosis. This innovative approach holds great potential for improving breast cancer screening and, thereby, enhancing our understanding of the disease and its detection patterns.

Multi-wavelength interference phase imaging for automatic breast cancer detection and delineation using diffuse reflection imaging.

Millions of women globally are impacted by the major health problem of breast cancer (BC). Early detection of BC is critical for successful treatment and improved survival rates. In this study, we provide a progressive approach for BC detection using multi-wavelength interference (MWI) phase imaging based on diffuse reflection hyperspectral (HS) imaging. The proposed findings are based on the measurement of the interference pattern between the blue (446.6 nm) and red (632 nm) wavelengths. We consider implementing a comprehensive image processing and categorization method based on the use of Fast Fourier (FF) transform analysis pertaining to a change in the refractive index between tumor and normal tissue. We observed that cancer growth affects tissue organization dramatically, as seen by persistently increased refractive index variance in tumors compared normal areas. Both malignant and normal tissue had different depth data collected from it that was analyzed. To enhance the categorization of ex-vivo BC tissue, we developed and validated a training classifier algorithm specifically designed for categorizing HS cube data. Following the application of signal normalization with the FF transform algorithm, our methodology achieved a high level of performance with a specificity (Spec) of 94% and a sensitivity (Sen) of 90.9% for the 632 nm acquired image categorization, based on preliminary findings from breast specimens under investigation. Notably, we successfully leveraged unstained tissue samples to create 3D phase-resolved images that effectively highlight the distinctions in diffuse reflectance features between cancerous and healthy tissue. Preliminary data revealed that our imaging method might be able to assist specialists in safely excising malignant areas and assessing the tumor bed following resection automatically at different depths. This preliminary investigation might result in an effective "in-vivo" disease description utilizing optical technology using a typical RGB camera with wavelength-specific operation with our quantitative phase MWI imaging methodology.

Delineation and detection of breast cancer using novel label-free fluorescence.

BACKGROUND: Accurate diagnosis of breast cancer (BC) plays a crucial role in clinical pathology analysis and ensuring precise surgical margins to prevent recurrence. METHODS: Laser-induced fluorescence (LIF) technology offers high sensitivity to tissue biochemistry, making it a potential tool for noninvasive BC identification. In this study, we utilized hyperspectral (HS) imaging data of stimulated BC specimens to detect malignancies based on altered fluorescence characteristics compared to normal tissue. Initially, we employed a HS camera and broadband spectrum light to assess the absorbance of BC samples. Notably, significant absorbance differences were observed in the 440-460 nm wavelength range. Subsequently, we developed a specialized LIF system for BC detection, utilizing a low-power blue laser source at 450 nm wavelength for ten BC samples. RESULTS: Our findings revealed that the fluorescence distribution of breast specimens, which carries molecular-scale structural information, serves as an effective marker for identifying breast tumors. Specifically, the emission at 561 nm exhibited the greatest variation in fluorescence signal intensity for both tumor and normal tissue, serving as an optical predictive biomarker. To enhance BC identification, we propose an advanced image classification technique that combines image segmentation using contour mapping and K-means clustering (K-mc, K = 8) for HS emission image data analysis. CONCLUSIONS: This exploratory work presents a potential avenue for improving "in-vivo" disease characterization using optical technology, specifically our LIF technique combined with the advanced K-mc approach, facilitating early tumor diagnosis in BC.

Emerging technology for intraoperative margin assessment and post-operative tissue diagnosis for breast-conserving surgery.

INTRODUCTION: Tissue-preserving surgery is utilized progressively in cancer therapy, where a clear surgical margin is critical to avoid cancer recurrence, specifically in breast cancer (BC) surgery. The Intraoperative pathologic approaches that rely on tissue segmenting and staining have been recognized as the ground truth for BC diagnosis. Nevertheless, these methods are constrained by its complication and timewasting for tissue preparation. OBJECTIVE: We present a non-invasive optical imaging system incorporating a hyperspectral (HS) camera to discriminate between cancerous and non-cancerous tissues in ex-vivo breast specimens, which could be an intraoperative diagnostic technique to aid surgeons during surgery and later a valuable tool to assist pathologists. METHODS: We have established a hyperspectral Imaging (HSI) system comprising a push-broom HS camera at wavelength 380∼1050 nm with source light 390∼980 nm. We have measured the investigated samples' diffuse reflectance (Rd), fixed on slides from 30 distinct patients incorporating mutually normal and ductal carcinoma tissue. The samples were divided into two groups, stained tissues during the surgery (control group) and unstained samples (test group), both captured with the HSI system in the visible and near-infrared (VIS-NIR) range. Then, to address the problem of the spectral nonuniformity of the illumination device and the influence of the dark current, the radiance data were normalized to yield the radiance of the specimen and neutralize the intensity effect to focus on the spectral reflectance shift for each tissue. The selection of the threshold window from the measured Rd is carried out by exploiting the statistical analysis by calculating each region's mean and standard deviation. Afterward, we selected the optimum spectral images from the HS data cube to apply a custom K-means algorithm and contour delineation to identify the regular districts from the BC regions. RESULTS: We noticed that the measured spectral Rd for the malignant tissues of the investigated case studies versus the reference source light varies regarding the cancer stage, as sometimes the Rd is higher for the tumor or vice versa for the normal tissue. Later, from the analysis of the whole samples, we found that the most appropriate wavelength for the BC tissues was 447 nm, which was highly reflected versus the normal tissue. However, the most convenient one for the normal tissue was at 545 nm with high reflection versus the BC tissue. Finally, we implement a moving average filter for noise reduction and a custom K-means clustering algorithm on the selected two spectral images (447, 551 nm) to identify the various regions and effectively-identified spectral tissue variations with a sensitivity of 98.95%, and specificity of 98.44%. A pathologist later confirmed these outcomes as the ground truth for the tissue sample investigations. CONCLUSIONS: The proposed system could help the surgeon and the pathologist identify the cancerous tissue margins from the non-cancerous tissue with a non-invasive, rapid, and minimum time method achieving high sensitivity up to 98.95%.

Quantitative phase analysis and hyperspectral imaging for the automatic identification of veins and blood perfusion maps.

INTRODUCTION: Medical workers commonly physically identify subcutaneous veins to locate a suitable vesselto implant a catheter for drug administration or blood sample. The general rule of thumb is to locate a big, clean vein that will allow the medication to readily pass within the intended blood vessel. Peripheral problematic venous access happens when a patient's veins are difficult to palpate because of factors like dark skin tone, edema or excess tissue. The ability to see how the vasculature changes to support the therapeutic methods without damaging the surrounding tissue is another challenge. MATERIALS AND METHODS: Hyperspectral imaging (HI) is a developing technique with several potential uses in medicine. Using its spectroscopic data, veins and arterioles could be noninvasively detected and discriminated. It is frequently important to use quantitative phase analysis for vein localization. To assess hyperspectral image data for the detection of both veins and peripheral arteries, we suggest using an advanced image processing and classification algorithm based on the phase information related to the index of refraction change and associated scattering. We show that this need may be satisfied using quantitative phase imaging of forearm skin tissue at different depths. RESULTS: To demonstrate the variations in the diffuse reflectance characteristics between skin surface and veins, phase resolved pictures were successfully produced for twelve volunteers using our imaging methodology. We found that the skin surface details are completely apparent at the unique wavelength of 441 nm. The 500-nm wavelength was the most efficient for grouping peripheral arteries and illuminating the blood perfusion maps. Using our HI experimental setup and our phase imaging procedure on the 600 nm and 650 nm visible spectral pictures, we were able to properly describe the vein map. This spectral area may be utilized as a vein locator marker which could approximately reach till 3 mm depth under skin surface. CONCLUSIONS: Initial findings suggested that our imaging technique would be able to assist medical examiners in safely assessing the veins and arteriole's locations automatically without exposing the skin to infrared radiation. Meanwhile, our pilot research in this work to determine the best spectral wavelengths for localizing veins and mapping blood perfusion using our phase analysis imaging strategy with the HI camera. By substituting the HI camera with a custom conventional RGB camera that only functions at specific wavelengths during the discovering of blood perfusion locations or prior to intravenous catheterization, a distinctive and efficient system for precise identification may be developed to serve in the field of the vascular therapeutic methods.

Non-invasive caries detection and delineation via novel laser-induced fluorescence with hyperspectral imaging.

Carious is a global chronic disease; 2 billion people and 520 million children suffer from permanent and primary teeth caries respectively. Early caries detection via precise, non-invasive, non-ionizing radiation is highly appreciated. Carious deteriorate the chemical structure of sound tooth tissues, with variation in its optical properties. In this study, customized laser-induced fluorescence system consists of non-ionizing laser light source and hyperspectral camera was developed for early caries detection. Tested tooth sample was illuminated with laser source of 385 nm and 5 mW power. The emitted spectrum signature for main tooth elements including enamel, dentin, stain, and caries were captured. Logarithmic scale of spectrum signature was applied in an attempt to enhance system sensitivity to fluorescent signal. Fluorescence signature at 500 nm secured the maximum fluorescence intensity difference for different tooth elements. Consequently 2D hyperspectral image at 500 nm was constructed. Enhanced 2D image was accomplished via nonlinear filter to enhance contrast. Segmentation via K mean clustering was adopted for precise caries delineation. This narrative, facile, non-invasive, non-ionizing technique experienced precise and accurate delineation of different caries stages (normal, moderate, and severe).

Hyperspectral image-based analysis of thermal damage for ex-vivo bovine liver utilizing radiofrequency ablation.

BACKGROUND & OBJECTIVE: Thermal ablation is the predominant methodology to treat liver tumors for segregating patients who are not permitted to have surgical intervention. However, noticing or predicting the size of the thermal strategies is a challenging endeavor. We aim to analyze the effects of ablation district volume following radiofrequency ablation (RFA) of ex-vivo liver exploiting a custom Hyperspectral Imaging (HSI) system. MATERIALS AND METHODS: RFA was conducted on the ex-vivo bovine liver at focal and peripheral blood vessel sites and observed by Custom HSI system, which has been designed to assess the exactness and proficiency using visible and near-infrared wavelengths region for tissue thermal effect. The experiment comprised up to ten trials with RFA. The experiment was carried out in two stages to assess the percentage of the thermal effect on the investigated sample superficially and for the side penetration effect. Measuring the diffuse reflectance (Rd) of the sample to identify the spectral reflectance shift which could differentiate between normal and ablated tissue exploiting the designed cross-correlation algorithm for monitoring of thermal ablation. RESULTS: Determination of the diffuse reflection (Rd) spectral signature responses from normal, thermal effected, and thermal ablation regions of the investigated liver sample. Where the ideal wavelength range at (600-640 nm) could discriminate between these different regions. Then, exploited the converted RGB image of the HS liver tissue after RFA for more validations which shows that the optimum wavelength for differentiation at (530-560 nm and 600-640 nm). Finally, applying statistical analysis to validate our results presenting that wavelength 600 nm had the highest standard deviation (δ) to differentiate between various thermally affected regions regarding the normal tissue and wavelength 640 nm shows the highest (δ) to differentiate between the ablated and normal regions. CONCLUSION: The designed and implemented medical imaging system incorporated the hyperspectral camera capabilities with the associate cross-correlation algorithm that could successfully distinguish between the ablated and thermally affected regions to assist the surgery during the tumor therapy.

Integration of transcriptomes analysis with spectral signature of total RNA for generation of affordable remote sensing of Hepatocellular carcinoma in serum clinical specimens.

Hepatocellular carcinoma (HCC) is a major global health problem with about 841,000 new cases and 782,000 deaths annually, due to lacking early biomarker/s, and centralized diagnosis. Transcriptomes research despite its infancy has proved excellence in its implementation in identifying a coherent specific cancer RNAs differential expression. However, results are sometimes overlapped by other cancer types which negatively affecting specificity, plus the high cost of the equipment used. Hyperspectral imaging (HSI) is an advanced tool with unique, spectroscopic features, is an emerging tool that has widely been used in cancer detection. Herein, a pilot study has been performed for HCC diagnosis, by exploiting HIS properties and the analysis of the transcriptome for the development of non-invasive remote HCC sensing. HSI data cube images of the sera extracted total RNA have been analyzed in HCC, normal subject, liver benign tumor, and chronic HCV with cirrhotic/non-cirrhotic liver groups. Data analyses have revealed a specific spectral signature for all groups and can be easily discriminated; at the computed optimum wavelength. Moreover, we have developed a simple setup based on a commercial laser pointer for sample illumination and a Smartphone CCD camera, with HSI consistent data output. We hypothesized that RNA differential expression and its spatial organization/folding are the key players in the obtained spectral signatures. To the best of our knowledge, we are the first to use HSI for sensing cancer based on total RNA in serum, using a Smartphone CCD camera/laser pointer. The proposed biosensor is simple, rapid (2 min), and affordable with specificity and sensitivity of more than 98% and high accuracy.

Total RNA nonlinear polarization: towards facile early diagnosis of breast cancer.

Different cancers are caused by accumulation of numerous genetic and epigenetic changes. Recently, nonlinear polarization has been considered as a marvelous tool for several medical applications. The capability of nonlinear polarization, to monitor any changes in RNA's spectral signature due to breast cancer (BC) was evaluated. Blood samples, from healthy controls and BC patients, were collected for whole blood preparation for genomic total RNA purification. Total RNA samples were stimulated with a light-emitting diode (LED) source of 565 nm; the resonance frequency of investigated RNA samples was captured and processed via hyperspectral imaging. Resonance frequency signatures were processed using fast Fourier transform in an attempt to differentiate between RNA (control) and RNA (BC) via frequency response. RNA (BC) demonstrated a characteristic signal at 0.02 GHz, as well as a phase shift at 0.031, and 0.070 GHZ from RNA (control). These features could offer early BC diagnosis. This is the first time to describe an optical methodology based on nonlinear polarization as a facile principle to distinguish and identify RNA alterations in BC by their characteristic fingerprint spectral signature.

Custom optical imaging system for ex-vivo breast cancer detection based on spectral signature.

BACKGROUND AND PURPOSE: Breast cancer is a popular well-known tumor in women globally and the subsequent driving reason for malignancy death. The purpose of the present study is to develop Low cost, commercial, and affordable system that discriminates malignant from normal breast tissues by exploiting the unique properties of Hyperspectral (HS) Imaging. MATERIALS AND METHODS: The difference in the optical properties of the investigated breast tissues gives various reactions to light transmission, absorption, and especially the reflection over the spectral range. A custom optical imaging system (COIS) was designed to assess variable responses to monochromatic LEDs (415, 565, 660 nm) to highlight the differences in the reflectance properties of malignant/normal tissue. Statistical analysis was computed for determining the ideal wavelength to differentiate between normal and malignant regions. The experiment was repeated using the same LEDs, and low-cost CCD camera to examine the capability of such a system to discriminate between normal and malignant tissue. RESULTS: Spectral images obtained by Hyperspectral camera, have been analyzed to reveal the difference of reflectance malignant and normal breast tissue. Superficial spectral reflection image with blue LED (415 nm) showed high variance (10.11). However, a more-depth reflection image with red LED (660 nm) showed low variance (4.44). So the optimum contrast image was produced by combining the three spectral information images from blue, green, and red LED. The COIS using a commercial CCD camera was in agreement with the HS camera. CONCLUSIONS: The novel COIS of the commercial Low-cost CCD Camera is reliable and can be used with endoscopy technique as an assistant tool for surgical doctor to make decision and assess the resection edges in real time during surgery.

Hyperspectral imaging for diagnosis and detection of ex-vivo breast cancer.

BACKGROUND AND PURPOSE: Breast cancer is one of the most widely recognized tumors. .Diagnosis made in the early stage of disease may imporve outcomes. The discovery of malignant growth utilizing noninvasive light intrusive methods in lieu of conventional excisional biopsy may assist in achieving this goal. MATERIALS AND METHODS: The change of the optical properties of ex-vivo breast tissues provides different responses to light transmission, absorption, and particularly the reflection over the spectrum range. We offer the use of Hyperspectral imaging (HSI) with advanced image processing and pattern recognition in order to analyze HSI data for breast cancer detection. The spectral signatures were mined and evaluated in both malignant and normal tissue. K-mean clustering was designed for classifying hyperspectral data in order to evaluate and detection of cancer tissue. This method was used to detect ex-vivo breast cancer. Spatial spectral images were created to high spot the differences in the reflectance properties of malignant versus normal tissue. RESULTS: Trials showed that the superficial spectral reflection images within 500 nm wavelength showed high variance (214.65) between cancerous and normal breast tissues. On the other hand, image within 620 nm wavelength showed low variance (0.0020).However, the superimposed of spectral region 420-620 nm was proposed as the optimum bandwidth. Finally, the proposed HS imaging system was capable to discriminate the tumor region from normal tissue of the ex-vivo breast sample with sensitivity and a specificity of 95 % and 96 %. CONCLUSIONS: High sensitivity and specificity were achieved, which proposes potential for HSI as an edge evaluation method to enhance the surgical outcome compared to the presently available techniques in the clinics.

Tissue characterization utilizing hyperspectral imaging for liver thermal ablation.

BACKGROUND: Thermal ablation is the dominant modality to treat liver tumors for patients who are not surgical candidates. . However, correctly predicting the volume of the subsequent tissue destruction during the Thermal Ablation technique is a difficult undertaking. OBJECTIVE: To examine the impacts of ablation zone volume following Radiofrequency ablation (RFA) of an ex-vivo bovine liver to correlate the impacts of thermal ablation with target organ perfusion; by exploiting the unique properties of Hyperspectral Imaging (HSI). MATERIALS AND METHODS: Radiofrequency ablation was perfused on ex-vivo bovine livers at peripheral and central­vessel­adjacent locations, and monitored by HSI with a spectral range from 400 to 1000 nm. The system contains k-means clustering (K = 8) algorithms combining spectral and spatial information. Labeled spectral signatures datasets were used as training data. Statistical analysis (10 samples) was computed to calculate the highest variance between six spectral images for determining the optimum wavelength for discrimination between the affected regions after thermal ablation (normal, thermal, and ablated liver tissue regions). RESULTS: The change of the optical properties ofex-vivo liver tissues provides different responses to light transmission, scattering, absorption and particularly the reflection over the spectrum range. The produced spectral image from reflection with the highest variance (358.07) empowered us to determine the optimum wavelength spectral image (720 ±â€¯18.92 nm) to distinguish between the normal, ablated, and thermal categories. CONCLUSION: Hyperspectral imaging is a powerful tool in monitoring tissue characterization, which is a useful technique for edge evaluation of liver thermal ablation ..

High-performance near-infrared imaging for breast cancer detection.

We present a method for the noninvasive determination of the size, position, and optical properties of tumors in the human breast. The tumor is first detected by photothermal imaging. It is then sized, located, and optically characterized using designed digital image processing and edge-detection pattern recognition. The method assumes that the tumor is spherical and inhomogeneous and embedded in an otherwise homogeneous tissue. Heat energy is deposited in the tissue by absorption of near-infrared (NIR) Nd:YAG laser radiation, and its subsequent conversion to heat via vibrational relaxation causes a rise in temperature of the tissue. The tumor absorbs and scatters NIR light more strongly than the surrounding healthy tissue. Heat will diffuse through the tissue, causing a rise in temperature of the surrounding tissue. Differentiation between normal and cancerous tissues is determined using IR thermal imaging. Results are presented on a 55-year-old patient with a papillary breast cancer. We found that these results provide the clinician with more detailed information about breast lesions detected by photothermal imaging and thereby enhance its potential for specificity.