Three-dimensional maps of human skin properties on full face with shadows using 3-D hyperspectral imaging

Hyperspectral imaging for monitoring of free flaps of the oral cavity: A feasibility study

OBJECTIVES

Hyperspectral imaging (HSI) provides spectral information about hemoglobin, water and oxygen supply and has thus great potential in perfusion monitoring. The aim of the present study was to investigate the feasibility of HSI in the postoperative monitoring of intraoral free flaps.

METHODS

The 14 patients receiving reconstructive head and neck surgery with a radial forearm free flap were included. HSI was performed intraoperatively (t0), on Day 1 (t1), 2 (t2), 3-6 (t3), 7-9 (t4), 10-11 (t5) and 12-15 (t6) postoperatively. Flap tissue perfusion was assessed on defined regions of interest by calculating the perfusion indices Tissue Hemoglobin Index (THI), hemoglobin oxygenation (StO2 ), Near Infrared Perfusion Index (NIR Perfusion Index) and Tissue Water Index (TWI).

RESULTS

Image quality varied depending on location of the flap and time of measurement. StO2 was >50 intraoperatively and >40 on t1 for all patients. A significant difference was found solely for TWI between t0 and t2 and t0 and t4. No flap loss occurred.

CONCLUSIONS

The use of HSI in the monitoring of intraoral flaps is feasible and might become a valuable addition to the current clinical examination of free flaps.

Measuring changes in blood volume fraction during induced gingivitis of healthy and unhealthy populations using hyperspectral spatial frequency domain imaging: a clinical study

This investigation aimed to quantitatively measure the changes in inflammation of subjects with healthy and unhealthy gums during a period of induced gingivitis. A total of 30 subjects (15 healthy, 15 with gum inflammation) were enlisted and given oral exams by a dental hygienist. Baseline measurements were acquired before a 3-week period of oral hygiene abstinence. The lobene modified gingival index scoring was used for inflammation scoring and hyperspectral spatial frequency domain imaging was used to quantitatively measure oxy- and deoxygenated blood volume fraction at two time points: at Baseline and after 3 weeks of oral hygiene abstinence. We found that abstaining from oral hygiene causes a near proportional increase in oxygenated and deoxygenated blood volume fraction for healthy individuals. For individuals who started the study with mild to moderate gingivitis, increases in blood volume were mainly due to deoxygenated blood.

Effects of Artificial Extraoral Markers on Accuracy of Three-Dimensional Dentofacial Image Integration: Smartphone Face Scan versus Stereophotogrammetry

Recently, three-dimensional (3D) facial scanning has been gaining popularity in personalized dentistry. Integration of the digital dental model into the 3D facial image allows for a treatment plan to be made in accordance with the patients’ individual needs. The aim of this study was to evaluate the effects of extraoral markers on the accuracy of digital dentofacial integrations. Facial models were generated using smartphone and stereophotogrammetry. Dental models were generated with and without extraoral markers and were registered to the facial models by matching the teeth or markers (n = 10 in each condition; total = 40). Accuracy of the image integration was measured in terms of general 3D position, occlusal plane, and dental midline deviations. The Mann−Whitney U test and two-way analysis of variance were used to compare results among face-scanning systems and matching methods (α = 0.05). As result, the accuracy of dentofacial registration was significantly affected by the use of artificial markers and different face-scanning systems (p < 0.001). The deviations were smallest in stereophotogrammetry with the marker-based matching and highest in smartphone face scans with the tooth-based matching. In comparison between the two face-scanning systems, the stereophotogrammetry generally produced smaller discrepancies than smartphones.

Multimodal hyperspectral fluorescence and spatial frequency domain imaging for tissue health diagnostics of the oral cavity

A multimodal, hyperspectral imaging system was built for diagnostics of oral tissues. The system, termed Hyperspectral-Fluorescence-Spatial Frequency Domain Imaging (Hy-F-SFDI), combines the principles of spatial frequency domain imaging, quantitative light fluorescence, and CIELAB color measurement. Hy-F-SFDI employs a compact LED projector, excitation LED, and a 16 channel hyperspectral camera mounted on a custom platform for tissue imaging. A two layer Monte Carlo approach was used to generate a reference table for quick tissue analysis. To demonstrate the clinical capabilities of Hy-F-SFDI, we used the system to quantify gingival tissue hemoglobin volume fraction, detect caries, bacterial activity, and measure tooth color of a volunteer at different time points. Hy-F-SFDI was able to measure quantitative changes in tissue parameters.

Fluorescence-advanced videodermatoscopy (FAV) for the differential diagnosis of suspicious facial lesions: A single-centre experience with pattern analysis and histopathological correlation

BACKGROUND

Fluorescence-advanced videodermatoscopy (FAV) is a new non-invasive high-resolution skin imaging technique to assess pigmented lesions in conjunction with the clinical examination and dermatoscopy.

OBJECTIVES

This is the first prospective study to identify morphologic descriptors and standardized terminology to examine facial pigmented lesions using FAV. The objectives were to identify FAV indicators, which can assist physicians in diagnosing suspicious flat facial pigmented lesions.

METHODS

Consecutive equivocal pigmented lesions were retrospective analysed. Histopathological examination was performed for all the lesions. The main cytomorphological and cytoarchitectural FAV features were described and correlated with histopathological characteristics.

RESULTS

From January to October 2020, 21 consecutive clinically suspected pigmented lesions in 20 patients were analysed using dermatoscopy and FAV and then surgically excised. Histopathological examination identified lentigo maligna (LM), lentigo maligna melanoma (LMM), solar lentigo (SL), flat seborrheic keratosis (SK) and pigmented actinic keratosis (PAK). Thirteen malignant melanocytic lesions were removed (11 LM, 2 LMM), two were diagnosed as PAK, and the remaining six pigmented lesions were SL-SKs. With FAV, large ovoid pleomorphic and dendritic cells arranged in the intrafollicular disposition, are typical of most malignant melanocytic lesions (12/13, 92.3%). No benign lesions displayed these features. In dermatoscopy, this folliculotropism corresponded to the presence of an annular-granular pattern with slate grey dots that were aggregated asymmetrically around follicular openings.

CONCLUSIONS

FAV features can provide an improved diagnostic approach in the differential diagnosis of flat pigmented facial lesions.

Accuracy of Portable Face-Scanning Devices for Obtaining Three-Dimensional Face Models: A Systematic Review and Meta-Analysis

The use of three-dimensional face-scanning systems to obtain facial models is of increasing interest, however, systematic assessments of the reliability of portable face-scan devices have not been widely conducted. Therefore, a systematic review and meta-analysis were performed considering the accuracy of facial models obtained by portable face-scanners in comparison with that of those obtained by stationary face-scanning systems. A systematic literature search was conducted in electronic databases following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for articles published from 1 January 2009 to 18 March 2020. A total of 2806 articles were identified, with 21 articles available for the narrative review and nine studies available for meta-analysis. The meta-analysis revealed that the accuracy of the digital face models generated by the portable scanners was not significantly different from that of the stationary face-scanning systems (standard mean difference (95% confidence interval) = -0.325 mm (-1.186 to 0.536); z = -0.74; p = 0.459). Within the comparison of the portable systems, no statistically significant difference was found concerning the accuracy of the facial models among scanning methods (p = 0.063). Overall, portable face-scan devices can be considered reliable for obtaining facial models. However, caution is needed when applying face-scanners with respect to scanning device settings, control of involuntary facial movements, landmark and facial region identifications, and scanning protocols.

Evaluating edge loss in the reflectance measurement of translucent materials

In many commercial instruments for measuring reflectance, the area illuminated on the measured object is identical to the area from which light is collected. This configuration is suitable for strongly scattering materials such as paper, but issues arise with translucent materials, because a portion of the incident light spreads around the illuminated area by subsurface transport and escapes the detection system. This phenomenon, referred to as edge loss, yields erroneous, underestimated reflectance measurements. In the case of colored and opalescent materials, the impact of edge loss on the measured reflectance varies with the wavelength, which is a significant issue for spectrophotometer and colorimeter users. In the present study, we investigate the edge-loss phenomenon with an emphasis on human skin measurement. In particular, we use a mathematical model to estimate the PSF of translucent materials, relying on the diffusion approximation of the radiative transfer theory, to predict edge-loss measurement error. We use this model to discuss the suitability of several commercial spectrophotometers to accurately measure the translucent materials of various optical properties and show that not all devices can adapt to all translucent materials.

Real-time skin chromophore estimation from hyperspectral images using a neural network

BACKGROUND

Hyperspectral imaging for in vivo human skin study has shown great potential by providing non-invasive measurement from which information usually invisible to the human eye can be revealed. In particular, maps of skin parameters including oxygen rate, blood volume fraction, and melanin concentration can be estimated from a hyperspectral image by using an optical model and an optimization algorithm. These applications, relying on hyperspectral images acquired with a high-resolution camera especially dedicated to skin measurement, have yielded promising results. However, the data analysis process is relatively expensive in terms of computation cost, with calculation of full-face skin property maps requiring up to 5 hours for 3-megapixels hyperspectral images. Such a computation time prevents punctual previewing and quality assessment of the maps immediately after acquisition.

METHODS

To address this issue, we have implemented a neural network that models the optimization-based analysis algorithm. This neural network has been trained on a set of hyperspectral images, acquired from 204 patients and their corresponding skin parameter maps, which were calculated by optimization.

RESULTS

The neural network is able to generate skin parameter maps that are visually very faithful to the reference maps much more quickly than the optimization-based algorithm, with computation times as short as 2 seconds for a 3-megapixel image representing a full face and 0.5 seconds for a 1-megapixel image representing a smaller area of skin. The average deviation calculated on selected areas shows the network's promising generalization ability, even on wide-field full-face images.

CONCLUSION

Currently, the network is adequate for preview purposes, providing relatively accurate results in a few seconds.

A background correction method to compensate illumination variation in hyperspectral imaging

Hyperspectral imaging (HSI) can measure both spatial (morphological) and spectral (biochemical) information from biological tissues. While HSI appears promising for biomedical applications, interpretation of hyperspectral images can be challenging when data is acquired in complex biological environments. Variations in surface topology or optical power distribution at the sample, encountered for example during endoscopy, can lead to errors in post-processing of the HSI data, compromising disease diagnostic capabilities. Here, we propose a background correction method to compensate for such variations, which estimates the optical properties of illumination at the target based on the normalised spectral profile of the light source and the measured HSI intensity values at a fixed wavelength where the absorption characteristics of the sample are relatively low (in this case, 800 nm). We demonstrate the feasibility of the proposed method by imaging blood samples, tissue-mimicking phantoms, and ex vivo chicken tissue. Moreover, using synthetic HSI data composed from experimentally measured spectra, we show the proposed method would improve statistical analysis of HSI data. The proposed method could help the implementation of HSI techniques in practical clinical applications, where controlling the illumination pattern and power is difficult.