Imaging Of Basal Cell Carcinoma Research Articles

Dermoscopy-based Radiomics Help Distinguish Basal Cell Carcinoma and Actinic Keratosis: A Large-scale Real-world Study Based on a 207-combination Machine Learning Computational Framework.

This study has used machine learning algorithms to develop a predictive model for differentiating between dermoscopic images of basal cell carcinoma (BCC) and actinic keratosis (AK). We compiled a total of 904 dermoscopic images from two sources - the public dataset (HAM10000) and our proprietary dataset from the First Affiliated Hospital of Dalian Medical University (DAYISET 1) - and subsequently categorised these images into four distinct cohorts. The study developed a deep learning model for quantitative analysis of image features and integrated 15 machine learning algorithms, generating 207 algorithmic combinations through random combinations and cross-validation. The final predictive model, formed by integrating XGBoost with Lasso regression, exhibited effective performance in the differential diagnosis of BCC and AK. The model demonstrated high sensitivity in the training set and maintained stable performance in three validation sets. The area under the curve (AUC) value reached 1.000 in the training set and an average of 0.695 in the validation sets. The study concludes that the constructed discriminative diagnostic model based on machine learning algorithms has excellent predictive capabilities that could enhance clinical decision-making efficiency, reduce unnecessary biopsies, and provide valuable guidance for further treatment.

A proposed architecture for convolutional neural networks to detect skin cancers

The goal of the research paper is to design and development of a computer-based system for the segmentation and classification of malignant skin diseases and a comparison between the accuracy of their detection, as two malignant diseases of skin diseases were detected. Namely, basal cell carcinoma and melanoma separately with images of nevus, and the images were collected from the ISIC 2020 archive group, as the total, The images used: 17,846 images include 3,008 images of basal cell carcinoma (BCC), 5,272 images of melanoma, and 9,566 images of a nevus, and validation data contains 20% of the images used which are not classified and randomly taken from the set of images, and the final test data contains 1,500 anonymous images. An architecture for the convolutional neural network technology in deep learning has been proposed that consists of a set of layers for processing. Processing raw input images for a group of pre-treatment transformations, the data augmentation process, so the number of images used became 86094 images of nevus, 27,072 images of BCC, and 47,448 images of melanoma. Through the detection process, the classification and detection accuracy of BCC was 98.25%, which is higher than the classification accuracy of melanoma is 91.61%.

Characterization of tissue types in basal cell carcinoma images via generative modeling and concept vectors.

The promise of machine learning methods to act as decision support systems for pathologists continues to grow. However, central to their successful adoption must be interpretable implementations so that people can trust and learn from them effectively. Generative modeling, most notable in the form of adversarial generative models, is a naturally interpretable technique because the quality of the model is explicit from the quality of images it generates. Such a model can be further assessed by exploring its latent space, using human-meaningful concepts by defining concept vectors. Motivated by these ideas, we apply for the first time generative methods to histological images of basal cell carcinoma (BCC). By simultaneously learning to generate and encode realistic image patches, we extract feature rich latent vectors that correspond to various tissue morphologies, namely BCC, epidermis, keratin, papillary dermis and inflammation. We show that a logistic regression model trained on these latent vectors can achieve high classification accuracies across 6 binary tasks (86-98%). Further, by projecting the latent vectors onto learned concept vectors we can generate a score for the absence or degree of presence for a given concept, providing semantically accurate "conceptual summaries" of the various tissues types within a patch. This can be extended to generate multi-dimensional heat maps for whole-image specimens, which characterizes the tissue in a similar way to a pathologist. We additionally find that accurate concept vectors can be defined using a small labeled dataset.

Assistant Diagnosis of Basal Cell Carcinoma and Seborrheic Keratosis in Chinese Population Using Convolutional Neural Network.

Objectives To evaluate CNN models' performance of identifying the clinical images of basal cell carcinoma (BCC) and seborrheic keratosis (SK) and to compare their performance with that of dermatologists. Methods We constructed a Chinese skin diseases dataset which includes 1456 BCC and 1843 SK clinical images and the corresponding medical history. We evaluated the performance using four mainstream CNN structures and transfer learning techniques. We explored the interpretability of the CNN model and compared its performance with that of 21 dermatologists. Results The fine-tuned InceptionResNetV2 achieved the best performance, with an accuracy and area under the curve of 0.855 and 0.919, respectively. Further experimental results suggested that the CNN model was not only interpretable but also had a performance comparable to that of dermatologists. Conclusions This study is the first on the assistant diagnosis of BCC and SK based on the proposed dataset. The promising results suggested that CNN model's performance was comparable to that of expert dermatologists.

Clinical image identification of basal cell carcinoma and pigmented nevi based on convolutional neural network

To construct an intelligent assistant diagnosis model based on the clinical images of basal cell carcinoma (BCC) and pigmented nevi in Chinese by using the advanced convolutional neural network (CNN). Methods: Based on the Xiangya Medical Big Data Platform, we constructed a large-scale clinical image dataset of skin diseases according to Chinese ethnicity and the Xiangya Skin Disease Dataset. We evaluated the performance of 5 mainstream CNN models (ResNet50, InceptionV3, InceptionResNetV2, DenseNet121, and Xception) on a subset of BCC and pigmented nevi of this dataset. We also analyzed the basis of the diagnosis results in the form of heatmaps. We compared the optimal CNN classification model with 30 professional dermatologists. Results: The Xiangya Skin Disease Dataset contains 150 223 clinical images with lesion annotations, covering 543 skin diseases, and each image in the dataset contains support for pathological gold standards and the patient's overall medical history. On the test set of 349 BCC and 497 pigmented nevi, the optimal CNN model was Xception, and its classification accuracy can reach 93.5%, of which the area under curve (AUC) values were 0.974 and 0.969, respectively. The results of the heatmap showed that the CNN model can indeed learn the characteristics associated with disease identification. The ability of the Xception model to identify clinical images of BCC and Nevi was basically comparable to that of professional dermatologists. Conclusion: This study is the first assistant diagnosis study for skin tumor based on Chinese ethnic clinical dataset. It proves that CNN model has the ability to distinguish between Chinese ethnicity's BCC and Nevi, and lays a solid foundation for the following application of artificial intelligence in the diagnosis and treatment for skin tumors.

Rapid histopathological imaging of skin and breast cancer surgical specimens using immersion microscopy with ultraviolet surface excitation

Rapid histopathological evaluation of fresh, unfixed human tissue using optical sectioning microscopy would have applications to intraoperative surgical margin assessment. Microscopy with ultraviolet surface excitation (MUSE) is a low-cost optical sectioning technique using ultraviolet illumination which limits fluorescence excitation to the specimen surface. In this paper, we characterize MUSE using high incident angle, water immersion illumination to improve sectioning. Propidium iodide is used as a nuclear stain and eosin yellow as a counterstain. Histologic features of specimens using MUSE, nonlinear microscopy (NLM) and conventional hematoxylin and eosin (H&E) histology were evaluated by pathologists to assess potential application in Mohs surgery for skin cancer and lumpectomy for breast cancer. MUSE images of basal cell carcinoma showed high correspondence with frozen section H&E histology, suggesting that MUSE may be applicable to Mohs surgery. However, correspondence in breast tissue between MUSE and paraffin embedded H&E histology was limited due to the thicker optical sectioning in MUSE, suggesting that further development is needed for breast surgical applications. We further demonstrate that the transverse image resolution of MUSE is limited by the optical sectioning thickness and use co-registered NLM to quantify the improvement in MUSE optical sectioning from high incident angle water immersion illumination.

628 Dual vibration resonance frequency CARS microscopy imaging of basal cell carcinoma to achieve stain free histopathology

628 Dual vibration resonance frequency CARS microscopy imaging of basal cell carcinoma to achieve stain free histopathology

Optical coherence tomography imaging of basal cell carcinoma undergoing photodynamic therapy: A pilot study

• The conversion of hyporeflective basal cell carcinoma nodules into hyperreflective ovoid structures was remarked with optical coherence tomography (OCT) as a sign of tumor regression under photodynamic therapy (PDT).

Autofluorescence imaging of basal cell carcinoma by smartphone RGB camera.

The feasibility of smartphones for in vivo skin autofluorescence imaging has been investigated. Filtered autofluorescence images from the same tissue area were periodically captured by a smartphone RGB camera with subsequent detection of fluorescence intensity decreasing at each image pixel for further imaging the planar distribution of those values. The proposed methodology was tested clinically with 13 basal cell carcinoma and 1 atypical nevus. Several clinical cases and potential future applications of the smartphone-based technique are discussed.

Texture analysis and classification in coherent anti-Stokes Raman scattering (CARS) microscopy images for automated detection of skin cancer

Coherent anti-Stokes Raman scattering (CARS) microscopy is a powerful tool for fast label-free tissue imaging, which is promising for early medical diagnostics. To facilitate the diagnostic process, automatic image analysis algorithms, which are capable of extracting relevant features from the image content, are needed. In this contribution we perform an automated classification of healthy and tumor areas in CARS images of basal cell carcinoma (BCC) skin samples. The classification is based on extraction of texture features from image regions and subsequent classification of these regions into healthy and cancerous with a perceptron algorithm. The developed approach is capable of an accurate classification of texture types with high sensitivity and specificity, which is an important step towards an automated tumor detection procedure.

A visual model approach to extract regions of interest in microscopical images of basal cell carcinoma

During the exploration of a microscopical slide the pathologist focuses the analyses on a few number of regions of interest. An accurately recognition of those regions may be a potential source of knowledge in many diagnostic task. However, due to the intrinsic complexity of the histopathological images, it is not a straightforward task. The method herein proposed is based on the cognitive process of looking for structures of interest.

Imaging of basal cell carcinoma by high-definition optical coherence tomography: histomorphological correlation. A pilot study

With the continued development of noninvasive therapies for basal cell carcinoma (BCC) such as photodynamic therapy and immune therapies, noninvasive diagnosis and monitoring become increasingly relevant. High-definition optical coherence tomography (HD-OCT) is a high-resolution imaging tool, with micrometre resolution in both transversal and axial directions, enabling visualization of individual cells up to a depth of around 570 μm, and filling the imaging gap between conventional optical coherence tomography (OCT) and reflectance confocal microscopy (RCM). We sought to determine the feasibility of detecting BCC by this technique using criteria defined for RCM and conventional OCT and compared with histology. In this pilot study skin lesions of 21 patients with a histologically proven BCC were imaged by HD-OCT just before excision and images analysed qualitatively. Features for four different BCC subtypes were described in both transverse and axial directions. In general, these features were subepidermal or intradermal aggregations of cells. These islands or trabeculae were surrounded by a less refractile border corresponding with palisading and peritumoral mucin production. There was a pronounced architectural disarray of the epidermis. A variably refractile stroma together with abundant dilated peritumoral blood vessels was present. These features were comparable with histological features for each patient. Using features already suggested by RCM and conventional OCT, the study implies that HD-OCT facilitates in vivo diagnosis of BCC and allows the distinction between different BCC subtypes for increased clinical utility.

High‐resolution imaging of basal cell carcinoma: a comparison between multiphoton microscopy with fluorescence lifetime imaging and reflectance confocal microscopy

The aim of this study was to compare morphological aspects of basal cell carcinoma (BCC) as assessed by two different imaging methods: in vivo reflectance confocal microscopy (RCM) and multiphoton tomography with fluorescence lifetime imaging implementation (MPT-FLIM). The study comprised 16 BCCs for which a complete set of RCM and MPT-FLIM images were available. The presence of seven MPT-FLIM descriptors was evaluated. The presence of seven RCM equivalent parameters was scored in accordance to their extension. Chi-squared test with Fisher's exact test and Spearman's rank correlation coefficient were determined between MPT-FLIM scores and adjusted-RCM scores. MPT-FLIM and RCM descriptors of BCC were coupled to match the descriptors that define the same pathological structures. The comparison included: Streaming and Aligned elongated cells, Streaming with multiple directions and Double alignment, Palisading (RCM) and Palisading (MPT-FLIM), Typical tumor islands, and Cell islands surrounded by fibers, Dark silhouettes and Phantom islands, Plump bright cells and Melanophages, Vessels (RCM), and Vessels (MPT-FLIM). The parameters that were significantly correlated were Melanophages/Plump Bright Cells, Aligned elongated cells/Streaming, Double alignment/Streaming with multiple directions, and Palisading (MPT-FLIM)/Palisading (RCM). According to our data, both methods are suitable to image BCC's features. The concordance between MPT-FLIM and RCM is high, with some limitations due to the technical differences between the two devices. The hardest difficulty when comparing the images generated by the two imaging modalities is represented by their different field of view.

Multiphoton laser tomography and fluorescence lifetime imaging of basal cell carcinoma: morphologic features for non‐invasive diagnostics

Multiphoton laser tomography (MPT) combined with fluorescence lifetime imaging (FLIM) is a non-invasive imaging technique, which gives access to the cellular and extracellular morphology of the skin. The aim of our study was to assess the sensitivity and specificity of MPT/FLIM descriptors for basal cell carcinoma (BCC), to improve BCC diagnosis and the identification of tumor margins. In the preliminary study, FLIM images referring to 35 BCCs and 35 healthy skin samples were evaluated for the identification of morphologic descriptors characteristic of BCC. In the main study, the selected parameters were blindly evaluated on a test set comprising 63 BCCs, 63 healthy skin samples and 66 skin lesions. Moreover, FLIM values inside a region of interest were calculated on 98 healthy skin and 98 BCC samples. In the preliminary study, three epidermal descriptors and 7 BCC descriptors were identified. The specificity of the diagnostic criteria versus 'other lesions' was extremely high, indicating that the presence of at least one BCC descriptor makes the diagnosis of 'other lesion' extremely unlikely. FLIM values referring to BCC cells significantly differed from those of healthy skin. In this study, we identified morphological and numerical descriptors enabling the differentiation of BCC from other skin disorders and its distinction from healthy skin in ex vivo samples. In future, MPT/FLIM may be applied to skin lesions to provide direct clinical guidance before biopsy and histological examination and for the identification of tumor margins allowing a complete surgical removal.

Discrimination of basal cell carcinoma from benign lesions based on extraction of ulcer features in polarized‐light dermoscopy images

Ulcers are frequently visible in magnified, cross-polarized, dermoscopy images of basal cell carcinoma. An ulcer without a history of trauma, a so-called 'atraumatic' ulcer, is an important sign of basal cell carcinoma, the most common skin cancer. Distinguishing such ulcers from similar features found in benign lesions is challenging. In this research, color and texture features of ulcers are analyzed to discriminate basal cell carcinoma from benign lesions. Ulcers in polarized-light dermoscopy images of 49 biopsy-proven basal cell carcinomas were identified and manually selected. For 153 polarized-light dermoscopy images of benign lesions, those areas that most closely mimicked ulcers were similarly selected. Fifteen measures were analyzed over the areas of ulcers and ulcer mimics. Six of those measures were texture measures: energy, variance, smoothness, skewness, uniformity, and entropy. Nine of those measures were color measures: relative measures of red, green, and blue; chromaticity of red, green, and blue; and the ratios blue-to-green, blue-to-red, and green-to-red. A back-propagation artificial neural network was able to discriminate most of the basal cell carcinoma from benign lesions, with an area under the ROC curve as high as 92.46%, using color and texture features of ulcer areas. Separation of basal cell carcinoma from benign cutaneous lesions using image analysis techniques applied to ulcers is feasible. As ulcers are a critical feature in malignant lesions, this finding may have application in the automatic detection of basal cell carcinoma.

A supervised visual model for finding regions of interest in basal cell carcinoma images

This paper introduces a supervised learning method for finding diagnostic regions of interest in histopathological images. The method is based on the cognitive process of visual selection of relevant regions that arises during a pathologist's image examination. The proposed strategy emulates the interaction of the visual cortex areas V1, V2 and V4, being the V1 cortex responsible for assigning local levels of relevance to visual inputs while the V2 cortex gathers together these small regions according to some weights modulated by the V4 cortex, which stores some learned rules. This novel strategy can be considered as a complex mix of "bottom-up" and "top-down" mechanisms, integrated by calculating a unique index inside each region. The method was evaluated on a set of 338 images in which an expert pathologist had drawn the Regions of Interest. The proposed method outperforms two state-of-the-art methods devised to determine Regions of Interest (RoIs) in natural images. The quality gain with respect to an adaptated Itti's model which found RoIs was 3.6 dB in average, while with respect to the Achanta's proposal was 4.9 dB.

Semitranslucency in Dermoscopic Images of Basal Cell Carcinoma

The clinical feature Semitranslucency, defined as a semiliquid or jellylike appearance, is a useful optical phenomenon for the identification of basal cell carcinoma (BCC).1 Semitranslucent areas in dermoscopic images of BCC also display the smooth, jellylike appearance of the clinical counterpart. The color ranges from reddish pink, usually seen in the thickest areas, to dull orange, often in a peripheral location (Figure 1), to gray (Figure 2 and Figure 3, insets). Figures 1 through ​through33 are noncontact polarized dermoscopic images of BCC, and Figure 4 is the histopathologic view of Figure 3. Figure 1 Figure 2 Figure 3 Figure 4 Semitranslucency, which is easily appreciated without visual aids in larger BCCs, can be best appreciated with non-contact polarized dermoscopic images. Lighter colors indermoscopic images, such as those appearing with the semitranslucency phenomenon, have previously been shown to be best preserved with non contact polarized dermoscopy.2 The typical color is absent in contact polarized images, but the contrast in smoothness with the surrounding areas, the most important aspect of the jelly like phenomenon, still allows appreciation of semitranslucency in some contact polarized images (Figure 2, inset). The jellylike appearance may be lost in nonpolarized contact images (Figures 1 and ​and3,3, insets). Histopathologic examination of 10 BCCs showed that areas of semitranslucency possess basaloid tumor nodules close to the surface, with a diminished epidermal thickness and a diminished collagen layer (Figure 4). This histopathologic constellation allows transmission of light to an appreciable depth of the skin and reflection from within the tumor. Further studies of this optical phenomenon are needed to determine which histopathologic subtypes of BCC are best correlated with semitranslucency.

Development of Raman microspectroscopy for automated detection and imaging of basal cell carcinoma

We investigate the potential of Raman microspectroscopy (RMS) for automated evaluation of excised skin tissue during Mohs micrographic surgery (MMS). The main aim is to develop an automated method for imaging and diagnosis of basal cell carcinoma (BCC) regions. Selected Raman bands responsible for the largest spectral differences between BCC and normal skin regions and linear discriminant analysis (LDA) are used to build a multivariate supervised classification model. The model is based on 329 Raman spectra measured on skin tissue obtained from 20 patients. BCC is discriminated from healthy tissue with 90+/-9% sensitivity and 85+/-9% specificity in a 70% to 30% split cross-validation algorithm. This multivariate model is then applied on tissue sections from new patients to image tumor regions. The RMS images show excellent correlation with the gold standard of histopathology sections, BCC being detected in all positive sections. We demonstrate the potential of RMS as an automated objective method for tumor evaluation during MMS. The replacement of current histopathology during MMS by a "generalization" of the proposed technique may improve the feasibility and efficacy of MMS, leading to a wider use according to clinical need.

Multidimensional non-linear laser imaging of Basal Cell Carcinoma

We have used a multidimensional non-linear laser imaging approach to visualize ex-vivo samples of basal cell carcinoma (BCC). A combination of several non-linear laser imaging techniques involving fluorescence lifetime, multispectral two-photon and second-harmonic generation imaging has been used to image different skin layers. This approach has elucidated some morphological (supported by histopathological images), biochemical, and physiochemical differences of the healthy samples with respect to BCC ones. In particular, in comparison with normal skin, BCC showed a blue-shifted fluorescence emission, a higher fluorescence response at 800 nm excitation wavelength and a slightly longer mean fluorescence lifetime. Finally, the use of aminolevulinic acid as a contrast agent has been demonstrated to increase the constrast in tumor border detection. The results obtained provide further support for in-vivo non-invasive imaging of Basal Cell Carcinoma.

Terahertz Pulsed Spectroscopy of Human Basal Cell Carcinoma

Good contrast is seen between normal tissue and regions of tumor in terahertz pulsed imaging of basal cell carcinoma (BCC). To date, the source of contrast at terahertz frequencies is not well understood. In this paper we present results of a spectroscopy study comparing the terahertz properties (absorption coefficient and refractive index) of excised normal human skin and BCC. Both the absorption coefficient and refractive index were higher for skin that contained BCC. The difference was statistically significant over the range 0.2 to 2.0 THz (6.6 cm(-1) to 66.6 cm(-1)) for absorption coefficient and 0.25 to 0.90 THz (8.3 cm(-1) to 30 cm(-1)) for refractive index. The maximum difference for absorption was at 0.5 THz(16.7 cm(-1)). These changes are consistent with higher water content. These results account for the contrast seen in terahertz images of BCC and explain why parameters relating to the reflected terahertz pulse provide information about the lateral spread of the tumor. Knowing the properties of the tissue over the terahertz frequency range will enable the use of mathematical models to improve understanding of the terahertz response of normal and diseased tissue.