Emerging Technology for Intraoperative Margin and Assisting in Post-Surgery tissue diagnostic for Future Breast-Conserving

Deep learning-assisted multispectral imaging for early screening of skin diseases

INTRODUCTION

Melanocytic nevi (MN), warts, seborrheic keratoses (SK), and psoriasis are four common types of skin surface lesions that typically require dermatoscopic examination for definitive diagnosis in clinical dermatology settings. This process is labor-intensive and resource-consuming. Traditional methods for diagnosing skin lesions rely heavily on the subjective judgment of dermatologists, leading to issues in diagnostic accuracy and prolonged detection times.

OBJECTIVES

This study aims to introduce a multispectral imaging (MSI)-based method for the early screening and detection of skin surface lesions. By capturing image data at multiple wavelengths, MSI can detect subtle spectral variations in tissues, significantly enhancing the differentiation of various skin conditions.

METHODS

The proposed method utilizes a pixel-level mosaic imaging spectrometer to capture multispectral images of lesions, followed by reflectance calibration and standardization. Regions of interest were manually extracted, and the spectral data were subsequently exported for analysis. An improved one-dimensional convolutional neural network is then employed to train and classify the data.

RESULTS

The new method achieves an accuracy of 96.82% on the test set, demonstrating its efficacy.

CONCLUSION

This multispectral imaging approach provides a non-contact and non-invasive method for early screening, effectively addressing the subjective identification of lesions by dermatologists and the prolonged detection times associated with conventional methods. It offers enhanced diagnostic accuracy for a variety of skin lesions, suggesting new avenues for dermatological diagnostics.

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.

Custom Hyperspectral Imaging System Reveals Unique Spectral Signatures of Heart, Kidney, and Liver Tissues

The rapid advancement of diagnostic and therapeutic optical techniques for oncology demands a good understanding of the optical properties of biological tissues. This study explores the capabilities of hyperspectral (HS) cameras as a non-invasive and non-contact optical imaging system to distinguish and highlight spectral differences inbiological soft tissuesof three structures (kidney, heart, and liver) for use inendoscopic interventionoropen surgery. The study presents an optical system consisting of two individual setups, the transmission setup, and the reflection setup, both incorporating anHS camerawith apolychromatic light sourcewithin the range of 380 to 1050 nm to measure tissue's light transmission (Tr) and diffuse light reflectance (Rd), respectively. The optical system was calibrated with a customizedliquid optical phantom, then 30 samples from various organs were investigated fortissue characterizationby measuring both Tr and Rd at the visible and near infrared (VIS-NIR) band. We exploited the ANOVA test, subsequently by a Tukey's test on the created three independent clusters (kidney vs. heart: group I / kidney vs. liver: group II / heart vs. liver: group III) to identify the optimum wavelength for each tissue regarding their spectroscopic optical properties in the VIS-NIR spectrum. The optimum spectral span for the determined light Tr of the three groups was 640 ∼ 680 nm, and the ideal range was 720 ∼ 760 nm for the measured light Rd for mutual group I and group II. However, the group III range was different at a range of 520 ∼ 560 nm. Therefore, the investigation provides vital information concerning theoptimum spectral scalefor the computed light Tr and Rd of the investigatedbiological tissues(kidney, liver, and heart) to be employed in diagnostic andtherapeutic medical applications.

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.

Development of a custom optical imaging system for non-invasive monitoring and delineation of lower limb varicose veins using hyperspectral imaging and quantitative phase analysis

BACKGROUND

Varicose veins (VV) are a prevalent chronic venous disorder, particularly affecting women of childbearing age. This condition is associated with significant complications, including pain, discomfort, leg cramps, ulceration, reduced quality of life, absenteeism, and even mortality. This study aims to develop a custom non-invasive, non-contact optical imaging system combined with magnitude and phase image calculation to monitor and visualize varicose veins and their tributaries using hyperspectral imaging and quantitative phase analysis with a k-means clustering algorithm.

RESULTS

Ten volunteers participated in the optical imaging system study. They were exposed to a polychromatic light source spanning the wavelength range of 400 nm-950 nm. The diffuse reflection spectra for varicose veins exhibited a peak at 530 nm, while leg veins showed a peak at 780 nm. Hyperspectral images obtained at these specific wavelengths were normalized in order to homogenize the spectral signatures of each pixel (converting the hyperspectral image to 8 bit RGB image) and filtered using a moving average filter. Subsequently, the varicose veins and leg veins were delineated and detected using quantitative phase analysis and a k-means clustering algorithm.

CONCLUSION

In conclusion, the custom optical imaging system, utilizing hyperspectral imaging and the associated clustering algorithm, provides detailed information regarding the spatial distribution of varicose veins. This information can assist vascular physicians in facilitating easier diagnosis and treatment planning.

Analysis of Specimen Mammography with Artificial Intelligence to Predict Margin Status

BACKGROUND

Intraoperative specimen mammography is a valuable tool in breast cancer surgery, providing immediate assessment of margins for a resected tumor. However, the accuracy of specimen mammography in detecting microscopic margin positivity is low. We sought to develop an artificial intelligence model to predict the pathologic margin status of resected breast tumors using specimen mammography.

METHODS

A dataset of specimen mammography images matched with pathologic margin status was collected from our institution from 2017 to 2020. The dataset was randomly split into training, validation, and test sets. Specimen mammography models pretrained on radiologic images were developed and compared with models pretrained on nonmedical images. Model performance was assessed using sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC).

RESULTS

The dataset included 821 images, and 53% had positive margins. For three out of four model architectures tested, models pretrained on radiologic images outperformed nonmedical models. The highest performing model, InceptionV3, showed sensitivity of 84%, specificity of 42%, and AUROC of 0.71. Model performance was better among patients with invasive cancers, less dense breasts, and non-white race.

CONCLUSIONS

This study developed and internally validated artificial intelligence models that predict pathologic margins status for partial mastectomy from specimen mammograms. The models' accuracy compares favorably with published literature on surgeon and radiologist interpretation of specimen mammography. With further development, these models could more precisely guide the extent of resection, potentially improving cosmesis and reducing reoperations.

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.

Instant identification of dental white spot using K-means algorithm via laser-induced fluorescence and associated hyperspectral imaging

Dental caries (DC) is a chronic illness that affects 2 billion individuals worldwide, with 520 million of those suffering from it in their primary teeth. It is apparent that early DC detection and subsequent minimally invasive therapy are crucial clinical requirements. The first reversible clinical indication of demineralization is dental white spot (DWS) lesions. However, diagnosing DWS poses extreme challenges for practitioners. In this investigation, a customized laser-induced fluorescence system with a hyperspectral imaging (HI) camera and a non-ionization laser light supply was created for DWS localization and early DC detection. A UV laser diode source with a wavelength of 395 nm was used for light stimulation for the 10 test samples of teeth. The emitted signature of the main tooth components, including dentin, DWS, enamel, and DC, was recorded. An attempt was made to increase the system's sensitivity to the fluorescent signal by applying a logarithmic scale to the spectral signature. Moreover, further discrimination may be achieved by signal strength. We identified that the fluorescent signal's peak intensity at 771 nm works best for discriminating DWS from normal areas, or DC. For characterizing dentin, the re-emitted frequency at 500 nm has the maximum intensity. Next, we presented our imaging grouping strategy that combines visual enhancement through a moving average, MA, filtering and segmenting an image using K-means clustering (K-mc) (K = 8) for instant and precise DWS grouping for the constructed HI images at (500 nm and 771 nm). Despite the tiny structure and its DWS white appearance, our approach could successfully demarcate the DWS on the tested teeth. Dental examiners might benefit from our simple, non-invasive, non-ionizing optical diagnosis approach to help them make their first assessments and experience accurate and exact delineation of the DWS to obtain immediate and higher rates of early-stage DC detection.