Multispectral Fundus Photography of Choroidal Nevi With Trans-Palpebral Illumination

Purpose

The purpose of this study was to investigate the spectral characteristics of choroidal nevi and assess the feasibility of quantifying the basal diameter of choroidal nevi using multispectral fundus images captured with trans-palpebral illumination.

Methods

The study used a widefield fundus camera with multispectral (625 nm, 780 nm, 850 nm, and 970 nm) trans-palpebral illumination to examine eight subjects diagnosed with choroidal nevi. Geometric features of nevi, including border clarity, overlying drusen, and lesion basal diameter, were characterized. Clinical imagers, including scanning laser ophthalmoscopy (SLO), autofluorescence (AF), and optical coherence tomography (OCT), were utilized for comparative assessment.

Results

Fundus images depicted nevi as dark regions with high contrast against the background. Near-infrared (NIR) fundus images provided enhanced visibility of lesion borders compared to visible fundus images and SLO images. Lesion-background contrast measurements revealed 635 nm SLO at 11% and 625 nm fundus at 42%. Significantly enhanced contrasts were observed in NIR fundus images at 780 nm (73%), 850 nm (63%), and 970 nm (67%). For quantifying the diameter of nevi, NIR fundus images at 780 nm and 850 nm yielded a deviation of less than 10% when compared to OCT measurements.

Conclusions

NIR fundus photography with trans-palpebral illumination enhances nevi visibility and boundary definition compared to SLO. Agreement in diameter measurements with OCT validates the accuracy and reliability of this method for choroidal nevi assessment.

Translational Relevance

Multispectral fundus imaging with trans-palpebral illumination improves choroidal nevi visibility and accurately measures basal diameter, promising to enhance clinical practices in screening, diagnosis, and monitoring of choroidal nevi.

Melanoma clinicopathological groups characterized and compared with dermoscopy and reflectance confocal microscopy

BACKGROUND

Dermoscopic and reflectance confocal microscopy (RCM) correlations between morphologic groups of melanoma have not yet been described.

OBJECTIVE

Describe and compare dermoscopic and RCM features of cutaneous melanomas with histopathological confirmation.

METHODS

Single center, retrospective analysis of consecutive melanomas evaluated with RCM (2015-2019). Lesions were clinically classified as typical, nevus-like, amelanotic/nonmelanoma skin cancer (NMSC)-like, seborrheic keratosis (SK)-like and lentigo/lentigo maligna (LM)-like. Presence or absence of common facial and nonfacial melanoma dermoscopic and RCM patterns were recorded. Clusters were compared with typical lesions by multivariate logistic regression.

RESULTS

Among 583 melanoma lesions, significant differences between clusters were evident (compared to typical lesions). Observation of dermoscopic features (>50% of lesions) in amelanotic/NMSC-like lesions consistently displayed 3 patterns (atypical network, atypical vascular pattern + regression structures), and nevus-like and SK-like lesions and lentigo/LM-like lesions consistently displayed 2 patterns (atypical network + regression structures, and nonevident follicles + heavy pigmentation intensity). Differences were less evident with RCM, as almost all lesions were consistent with melanoma diagnosis.

LIMITATIONS

Small SK-like lesions sample, single RCM analyses (no reproduction of outcome).

CONCLUSION

RCM has the potential to augment our ability to consistently and accurately diagnose melanoma independently of clinical and dermoscopic features.

Understanding choroidal nevus risk factors for transformation into melanoma

A choroidal nevus is a common intraocular tumor in the United States, found in approximately 5% of Caucasian adults. The three main risks of melanocytic choroidal nevus include vision loss from a subfoveal nevus, development of subretinal fluid, and transformation of nevus into melanoma, a malignant counterpart. We explore clinical risk factors that predict benign melanocytic choroidal nevus transformation into a malignant choroidal melanoma. Based on a large analysis of 2,355 cases that were monitored longitudinally using multimodal imaging, the most recent list of clinical features includes tumor Thickness greater than 2 mm on ultrasonography, subretinal Fluid on optical coherence tomography, Symptomatic vision loss 20/50 or worse, Orange pigment on fundus autofluorescence, Melanoma hollow on ultrasonography, and DIaMeter greater than 5 mm on fundus photography. These factors are remembered with a mnemonic of the capital letters TFSOM-DIM for "To Find Small Ocular Melanoma Doing Imaging." Analysis of these factors demonstrated a Kaplan-Meier mean five-year risk of 1% with no risk factors, 11% with any one factor, 22% with any two factors, 34% with any three factors, 51% with any four factors, and 55% with any five factors. There was no patient with six risk factors. Of those with combinations of four risk factors, six of 15 combinations yielded a 70%-100% rate of transformation; of those with combinations of five risk factors, two of five combinations yielded a 70%-100% rate of transformation. Choroidal nevus carries a risk for evolving into melanoma, and understanding of clinical and imaging features predictive of this outcome is highly important.

Ultrasonographic findings in nevus lipomatosus cutaneous superficialis: What differentiates this tumor from other soft tissue tumors?

Nevus lipomatosus cutaneous superficialis is a rare, benign hamartoma characterized by mature adipocyte proliferation in the dermis. It is frequently difficult to distinguish clinically from soft tissue tumors, including lipoma, neurofibroma, venous malformation, and angiolipoma. Notably, the classical form, which shows multiple and sometimes enlarged nodules, is difficult to differentiate from liposarcoma based on clinical examination, computed tomography, and magnetic resonance imaging findings. Therefore, to ascertain the utility of ultrasonography in diagnosing nevus lipomatosus cutaneous superficialis, sonographic examinations were performed on eight patients with nevus lipomatosus cutaneous superficialis. All patients had ill-defined hyperechoic masses in the dermis or from the dermis to the subcutis, and the posterior echoes were attenuated in seven patients. Color Doppler sonography revealed no blood flow to the lesions. Ultrasound images were created using the reflections of ultrasound waves at interfaces with different acoustic impedances. Therefore, it is assumed that, in nevus lipomatosus cutaneous superficialis, the ultrasound beam is scattered by ectopic mature adipocytes intermingled with collagen bundles, which are shown as hyperechoic masses. Furthermore, the scattering of the ultrasound beam is thought to reduce tissue penetration, which may attenuate the posterior echo.

Polarization-Diversity Optical Coherence Tomography Assessment of Choroidal Nevi

Purpose

The purpose of this study was to demonstrate the utility of polarization-diversity optical coherence tomography (PD-OCT), a noninvasive imaging technique with melanin-specific contrast, in the quantitative and qualitative assessment of choroidal nevi.

Methods

Nevi were imaged with a custom-built 55-degree field-of-view (FOV) 400 kHz PD-OCT system. Imaging features on PD-OCT were compared to those on fundus photography, auto-fluorescence, ultrasound, and non-PD-OCT images. Lesions were manually segmented for size measurement and metrics for objective assessment of melanin distributions were calculated, including degree of polarization uniformity (DOPU), attenuation coefficient, and melanin occupancy rate (MOR).

Results

We imaged 17 patients (mean age = 69.5 years, range = 37-90) with 11 pigmented, 3 non-pigmented, and 3 mixed pigmentation nevi. Nevi with full margin acquisition had an average longest basal diameter of 5.1 mm (range = 2.99-8.72 mm) and average height of 0.72 mm (range = 0.37 mm-2.09 mm). PD-OCT provided clear contrast of choroidal melanin content, distribution, and delineation of nevus margins for melanotic nevi. Pigmented nevi were found to have lower DOPU, higher attenuation coefficient, and higher MOR than non-pigmented lesions. Melanin content on PD-OCT was consistent with pigmentation on fundus in 15 of 17 nevi (88%).

Conclusions

PD-OCT allows objective assessment of choroidal nevi melanin content and distribution. In addition, melanin-specific contrast by PD-OCT enables clear nevus margin delineation and may improve serial growth surveillance. Further investigation is needed to determine the clinical significance and prognostic value of melanin characterization by PD-OCT in the evaluation of choroidal nevi.

Refining skin lesions classification performance using geometric features of superpixels

This paper introduces superpixels to enhance the detection of skin lesions and to discriminate between melanoma and nevi without false negatives, in dermoscopy images. An improved Simple Linear Iterative Clustering (iSLIC) superpixels algorithm for image segmentation in digital image processing is proposed. The local graph cut method to identify the region of interest (i.e., either the nevi or melanoma lesions) has been adopted. The iSLIC algorithm is then exploited to segment sSPs. iSLIC discards all the SPs belonging to image background based on assigned labels and preserves the segmented skin lesions. A shape and geometric feature extraction task is performed for each segmented SP. The extracted features are fed into six machine learning algorithms such as: random forest, support vector machines, AdaBoost, k-nearest neighbor, decision trees (DT), Gaussian Naïve Bayes and three neural networks. These include Pattern recognition neural network, Feed forward neural network, and 1D Convolutional Neural Network for classification. The method is evaluated on the 7-Point MED-NODE and PAD-UFES-20 datasets and the results have been compared to the state-of-art findings. Extensive experiments show that the proposed method outperforms the compared existing methods in terms of accuracy.

Optical coherence tomography confirms non-malignant pigmented lesions in phacomatosis pigmentokeratotica using a support vector machine learning algorithm

INTRODUCTION

Phacomatosis pigmentokeratotica (PPK), an epidermal nevus syndrome, is characterized by the coexistence of nevus spilus and nevus sebaceus. Within the nevus spilus, an extensive range of atypical nevi of different morphologies may manifest. Pigmented lesions may fulfill the ABCDE criteria for melanoma, which may prompt a physician to perform a full-thickness biopsy.

MOTIVATION

Excisions result in pain, mental distress, and physical disfigurement. For patients with a significant number of nevi with morphologic atypia, it may not be physically feasible to biopsy a large number of lesions. Optical coherence tomography (OCT) is a non-invasive imaging modality that may be used to visualize non-melanoma and melanoma skin cancers.

MATERIALS AND METHOD

In this study, we used OCT to image pigmented lesions with morphologic atypia in a patient with PPK and assessed their quantitative optical properties compared to OCT cases of melanoma. We implement a support vector machine learning algorithm with Gabor wavelet transformation algorithm during post-image processing to extract optical properties and calculate attenuation coefficients.

RESULTS

The algorithm was trained and tested to extract and classify textural data.

CONCLUSION

We conclude that implementing this post-imaging machine learning algorithm to OCT images of pigmented lesions in PPK has been able to successfully confirm benign optical properties. Additionally, we identified remarkable differences in attenuation coefficient values and tissue optical characteristics, further defining separating benign features of pigmented lesions in PPK from malignant features.

Survey of ophthalmic imaging use to assess risk of progression of choroidal nevus to melanoma

BACKGROUND/OBJECTIVES

The aim of this study was to ascertain the use of ocular imaging and the updated screening criteria in the evaluation of choroidal nevus across the United States.

METHODS

Sixty ophthalmologists completed an anonymous 21-question survey addressing their use of the screening criteria for evaluating choroidal nevi, as well as their use of ultrasonography (US), optical coherence tomography (OCT), and autofluorescence (AF) in daily practice.

RESULTS

The majority of respondents were from the Northeast (55%), worked in private practice (83%), and practiced general ophthalmology (42%). The 2009 criteria TFSOM-UHHD was used by 39 (65%) respondents, while the 2019 criteria TFSOM-DIM was used by 29 (48%) respondents. Compared to anterior segment ophthalmologists, posterior segment ophthalmologists were more likely to use the TFSOM-UHHD criteria (94% vs. 53%, OR = 13.9, p = 0.014), the TFSOM-DIM criteria (88% vs. 33%, OR = 15.5, p < 0.001), fundus AF (82% vs. 19%, OR = 20.4, p < 0.001), and US (94% vs. 42%, OR = 22.2, p = 0.004) in daily practice.

CONCLUSIONS

From the survey of current practice patterns, we learned that there is a general trend of underutilization of the proper imaging modalities - and thus the criteria - in evaluating choroidal nevus. More education about ocular cancer and its screening could improve patient outcomes in the future.

Computer Aided Diagnosis of Melanoma Using Deep Neural Networks and Game Theory: Application on Dermoscopic Images of Skin Lesions

Early detection of melanoma remains a daily challenge due to the increasing number of cases and the lack of dermatologists. Thus, AI-assisted diagnosis is considered as a possible solution for this issue. Despite the great advances brought by deep learning and especially convolutional neural networks (CNNs), computer-aided diagnosis (CAD) systems are still not used in clinical practice. This may be explained by the dermatologist's fear of being misled by a false negative and the assimilation of CNNs to a "black box", making their decision process difficult to understand by a non-expert. Decision theory, especially game theory, is a potential solution as it focuses on identifying the best decision option that maximizes the decision-maker's expected utility. This study presents a new framework for automated melanoma diagnosis. Pursuing the goal of improving the performance of existing systems, our approach also attempts to bring more transparency in the decision process. The proposed framework includes a multi-class CNN and six binary CNNs assimilated to players. The players' strategies is to first cluster the pigmented lesions (melanoma, nevus, and benign keratosis), using the introduced method of evaluating the confidence of the predictions, into confidence level (confident, medium, uncertain). Then, a subset of players has the strategy to refine the diagnosis for difficult lesions with medium and uncertain prediction. We used EfficientNetB5 as the backbone of our networks and evaluated our approach on the public ISIC dataset consisting of 8917 lesions: melanoma (1113), nevi (6705) and benign keratosis (1099). The proposed framework achieved an area under the receiver operating curve (AUROC) of 0.93 for melanoma, 0.96 for nevus and 0.97 for benign keratosis. Furthermore, our approach outperformed existing methods in this task, improving the balanced accuracy (BACC) of the best compared method from 77% to 86%. These results suggest that our framework provides an effective and explainable decision-making strategy. This approach could help dermatologists in their clinical practice for patients with atypical and difficult-to-diagnose pigmented lesions. We also believe that our system could serve as a didactic tool for less experienced dermatologists.

Three-dimensional representation of triple spectral line imaging data as an option for noncontact skin diagnostics

SIGNIFICANCE

Skin malformations in dermatology are mostly evaluated subjectively, based on a doctor's experience and visual perception; an option for objective quantitative skin assessment is camera-based spectrally selective diagnostics. Multispectral imaging is a technique capable to provide information about concentrations of the absorbing chromophores and their distribution over the malformation in a noncontact way. Conversion of spectral images into distribution maps of chromophores can be performed by means of the modified Beer-Lambert law. However, such distribution maps represent only single specific cases, therefore, some extensive method for data comparison is needed.

AIM

This study aims to develop a more informative approach for identification and characterization of skin malformations using three-dimensional (3D) representation of triple spectral line imaging data.

APPROACH

The 3D-representation method is experimentally tested on eight different skin pathology types, including both benign and malignant pathologies; an imaging device ensuring uniform three laser line (448, 532, and 659 nm) illumination is used. Three spectral line images are extracted from a single snapshot RGB image data, with subsequent calculation of attenuation coefficients for each working wavelength at every image pixel and represented as 3D graphs. Skin chromophore content variations in malformations are represented in a similar way.

RESULTS

Clinical measurement results for 99 skin pathologies, including basal cell carcinomas, melanoma, dermal nevi, combined nevi, junctional nevi, blue nevi, seborrheic keratosis, and hemangiomas. They are presented as 3D spectral attenuation maps exhibiting specific individual features for each group of pathologies. Along with intensity attenuation maps, 3D maps for content variations of three main skin chromophores (melanin, oxyhemoglobin, and deoxyhemoglobin), calculated in frame of a model based on modified Beer-Lambert law, are also presented. Advantages and disadvantages of the proposed data representation method are discussed.

CONCLUSIONS

The described 3D-representation method of triple spectral line imaging data shows promising potential for objective quantitative noncontact diagnosis of skin pathologies.

Dermatoscopy and Reflective Confocal Microscopy for Facial Seborrheic Keratosis, Verruca Plana, and Nevus Pigmentosus

OBJECTIVE

To describe the features of facial seborrheic keratosis, verruca plana, and nevus pigmentosus by dermatoscopy and reflective confocal microscope (RCM).

STUDY DESIGN

Cross-sectional observational study.

PLACE AND DURATION OF STUDY

Dermatology Department of The Affiliated Hospital of Xuzhou Medical University from January 2017 to January 2019.

METHODOLOGY

Patients of either gender and age, clinically diagnosed as seborrheic keratosis, verruca plana, and nevus pigmentosus without any prior treatment, were enrolled. Patients with extremities and trunk involvement were excluded. One typical skin lesion was chosen from each patient and subjected to dermatoscopy and RCM separately; imaging features were recorded and analysed.

RESULTS

A total of 402 patients (183 men and 219 women) between 6 and 88 years of age (mean age 41.8 years) were inducted. The duration of disease was between one week and 10 years. Seborrheic keratosis on dermatoscopy presented as acne-like openings (122/172; 70.93%, milia-like cysts (113/172; 65.70%), hairpin-like vessels (108/172; 62.79%, brain-like structures (103/172; 59.88%, worm-like pharyngeal margins (17/172; 9.88%), and fingerprint-like structures (8/172; 4.65%). On RCM, it showed epidermal cerebral gyrus structure (165/172; 95.93%), superficial vasodilatation and hyperemia (81/172; 47.09%), and keratinous cysts (73/172; 42.44%). Verruca plana on dermatoscopy showed a number of punctate hemorrhages scattered against a light-red background (108/114; 94.74%); on RCM it showed rose-like concentric structures (89/114; 78.07%). The features of nevus pigmentosus observed by dermatoscopy were homogeneous mode (23/52; 44.23%), light brown color (30/52; 57.69%) for intradermal nevus; mesh mode (18/41; 43.90%, dark brown color, little black dots and spherical structure (both 18/41; 43.90%) were visible for junctional nevus; globular mode, reticular and cobblestone structures (both 11/23; 47.83%) for compound nevus. On RCM, there were a few nevus cells visible within the dermal papilla (52/52; 100%) for intradermal nevus; bright pebble-like structures accumulated in the basal layer, nested round and elliptical nevus cells (40/41; 97.56%) for junctional nevus; nevus cells in the epidermis and dermis (21/23; 91.30%) for compound nevus.

CONCLUSION

Used in combination with clinical manifestations, the application of dermatoscopy and RCM may help distinguish seborrheic keratosis, verruca plana, and nevus pigmentosus on the basis of their respective dermatoscopic and RCM features. Key Words: Dermatoscopy, Reflective confocal microscope, Seborrheic keratosis, Verruca plana, Nevus pigmentosus, Brain-like structures, Rose-like structure.

Ultrasound Imaging of Nevus Sebaceous of Jadassohn

Nevus sebaceous of Jadassohn (NSJ) is a cutaneous hamartoma commonly found in the scalp and face and more frequent in children. Clinically, it presents as a yellowish and hairless plaque. In later stages, the NSJ can develop secondary tumors, including skin cancer. We reviewed the ultrasound characteristics of 9 patients (67% female; mean age, 3 years) with NSJ at 18 and 70 MHz. The ultrasound analysis covers the data on the location, thickness, echo structure, and vascularity. The provision of the ultrasound patterns of NSJ can support early diagnosis, avoid unnecessary biopsies, and support monitoring.

Evaluation of 20-MHz high-frequency ultrasonography for the diagnosis of choroidal nevi

PURPOSE

Evaluate short-term intraoperator reproducibility of ultrasonographic measurements of choroidal nevi using 10- and 20-MHz probes, and the efficacy of the high-frequency probes for the diagnosis of choroidal nevi.

METHODS

Diameters and thicknesses of choroidal nevi were measured using a 10-MHz probe and a high-frequency long focal length 20-MHz probe (Quantel Medical™). The first part of the study evaluated intraoperator reproducibility of measurements of choroidal nevi with 10- and 20-MHz probes and the second part of the study allowed the comparisons of the measurements of largest tumor diameter (LDT) of choroidal nevi of 40 patients between the 10- and 20-MHz probes. The two-way random average agreement intraclass correlation coefficients (ICC), Bland-Altman plot, and a paired t test were used.

RESULTS

The intraoperator reproducibility of choroidal nevi measurements with 10- and 20-MHz probes was excellent (ICC > 0.9, n = 20). Four flat nevi, not detectable at 10 MHz, could be located with the high-frequency probe (p = 0.12). There was no significant difference in thickness or LTD measurements between the 10- and 20-MHz probes (n = 31). Both techniques showed an excellent agreement (ICC > 0.8) for thickness and LTD measurements. All the choroidal nevi that were not measurable with the 10-MHz probe (n = 7) were measured with the 20-MHz probe.

CONCLUSION

The high-frequency 20-MHz probe allows additional detection and measurements of flat choroidal nevi. When detectable, the ultrasonographic measurements of thickness and diameter of choroidal nevi are similar with both the 10- and the 20-MHz probes.

Microblotches on Dermoscopy of Melanocytic Lesions are Associated with Melanoma: A Cross-sectional Study

Numerous dermoscopic structures for the early detection of melanoma have been described. The aim of this study was to illustrate the characteristics of dermoscopic structures that are similar to blotches, but smaller (termed microblotches), and to evaluate their association with other well-known dermoscopic structures. A cross-sectional study design, including 165 dermoscopic images of melanoma was used to define microblotches, and 241 consecutive images of naevi from the HAM10000 database, were studied to evaluate the prevalence of this criterion in both groups. Microblotches were defined as sharply demarcated structures ≤1 mm, with geographical borders visible only with dermoscopy. Microblotches were present in 38.7% of the melanomas and 6.7% of the naevi. Moreover, microblotches were associated with an odds ratio (OR) of malignancy of 5.79, and were more frequent in invasive melanoma than in the in-situ subtype (OR 2.92). Histologically, they correspond to hyperpigmented parakeratosis or epidermal consumption. In conclusion, microblotches are related to melanomas. This finding could help dermatologists to differentiate between naevi and melanomas.

Semi-Supervised Nests of Melanocytes Segmentation Method Using Convolutional Autoencoders

In this research, we present a semi-supervised segmentation solution using convolutional autoencoders to solve the problem of segmentation tasks having a small number of ground-truth images. We evaluate the proposed deep network architecture for the detection of nests of nevus cells in histopathological images of skin specimens is an important step in dermatopathology. The diagnostic criteria based on the degree of uniformity and symmetry of border irregularities are particularly vital in dermatopathology, in order to distinguish between benign and malignant skin lesions. However, to the best of our knowledge, it is the first described method to segment the nests region. The novelty of our approach is not only the area of research, but, furthermore, we address a problem with a small ground-truth dataset. We propose an effective computer-vision based deep learning tool that can perform the nests segmentation based on an autoencoder architecture with two learning steps. Experimental results verified the effectiveness of the proposed approach and its ability to segment nests areas with Dice similarity coefficient 0.81, sensitivity 0.76, and specificity 0.94, which is a state-of-the-art result.

Integration of morphological preprocessing and fractal based feature extraction with recursive feature elimination for skin lesion types classification

BACKGROUND AND OBJECTIVE

Skin cancer is the commonest form of cancer in the worldwide population. Non-invasive and non-contact imaging modalities are being used for the screening of melanoma and other cutaneous malignancies to endorse early detection and prevention of the disease. Traditionally it has been a problem for medical personnel to differentiate melanoma, dysplastic nevi and basal cell carcinoma (BCC) diseases from one another due to the confusing appearance and similarity in the characteristics of the pigmented lesions. The paper reports an integrated method developed for identifying these skin diseases from the dermoscopic images.

METHODS

The proposed integrated computer-aided method has been employed for the identification of each of these diseases using recursive feature elimination (RFE) based layered structured multiclass image classification technique. Prior to the classification, different quantitative features have been extracted by analyzing the shape, the border irregularity, the texture and the color of the skin lesions, using different image processing tools. Primarily, a combination of gray level co-occurrence matrix (GLCM) and a proposed fractal-based regional texture analysis (FRTA) algorithm has been used for the quantification of textural information. The performance of the framework has been evaluated using a layered structure classification model using support vector machine (SVM) classifier with radial basis function (RBF).

RESULTS

The performance of the morphological skin lesion segmentation algorithm has been evaluated by estimating the pixel level sensitivity (Sen) of 0.9172, 0.9788 specificity (Spec), 0.9521 accuracy (ACU), along with the image similarity measuring indices as Jaccard similarity index (JSI) of 0.8562 and Dice similarity coefficient (DSC) of 0.9142 with respect to the corresponding ground truth (GT) images. The quantitative features extracted from the proposed feature extraction algorithms have been employed for the proposed multi-class skin disease identification. The proposed layered structure identifies all the three classes of skin diseases with a highly acceptable classification accuracy of 98.99%, 97.54% and 99.65% for melanoma, dysplastic nevi and BCC respectively.

CONCLUSION

To overcome the difficulties of proper diagnosis of diseases based on visual evaluation, the proposed integrated system plays an important role by quantifying the effective features and identifying the diseases with higher degree of accuracy. This combined approach of quantitative and qualitative analysis not only increases the diagnostic accuracy, but also provides some important information not obtainable from qualitative assessment alone.

Classification of skin lesions using transfer learning and augmentation with Alex-net

Skin cancer is one of most deadly diseases in humans. According to the high similarity between melanoma and nevus lesions, physicians take much more time to investigate these lesions. The automated classification of skin lesions will save effort, time and human life. The purpose of this paper is to present an automatic skin lesions classification system with higher classification rate using the theory of transfer learning and the pre-trained deep neural network. The transfer learning has been applied to the Alex-net in different ways, including fine-tuning the weights of the architecture, replacing the classification layer with a softmax layer that works with two or three kinds of skin lesions, and augmenting dataset by fixed and random rotation angles. The new softmax layer has the ability to classify the segmented color image lesions into melanoma and nevus or into melanoma, seborrheic keratosis, and nevus. The three well-known datasets, MED-NODE, Derm (IS & Quest) and ISIC, are used in testing and verifying the proposed method. The proposed DCNN weights have been fine-tuned using the training and testing dataset from ISIC in addition to 10-fold cross validation for MED-NODE and DermIS—DermQuest. The accuracy, sensitivity, specificity, and precision measures are used to evaluate the performance of the proposed method and the existing methods. For the datasets, MED-NODE, Derm (IS & Quest) and ISIC, the proposed method has achieved accuracy percentages of 96.86%, 97.70%, and 95.91% respectively. The performance of the proposed method has outperformed the performance of the existing classification methods of skin cancer.

Nonscanning large-area Raman imaging for ex vivo/in vivo skin cancer discrimination

Imaging Raman spectroscopy can be used to identify cancerous tissue. Traditionally, a step-by-step scanning of the sample is applied to generate a Raman image, which, however, is too slow for routine examination of patients. By transferring the technique of integral field spectroscopy (IFS) from astronomy to Raman imaging, it becomes possible to record entire Raman images quickly within a single exposure, without the need for a tedious scanning procedure. An IFS-based Raman imaging setup is presented, which is capable of measuring skin ex vivo or in vivo. It is demonstrated how Raman images of healthy and cancerous skin biopsies were recorded and analyzed.

Swept-Source OCT Angiography Identifies Choroidal Neovascularization Arising From a Choroidal Nevus

Swept-source optical coherence tomography angiography (SS-OCTA) was used to diagnose choroidal neovascularization (CNV) arising from a choroidal nevus. A 61-year-old woman initially presented with submacular hemorrhage. She was diagnosed with neovascular age-related macular degeneration (AMD) and received three injections of bevacizumab (Avastin; Genentech, South San Francisco, CA). At a follow-up visit, SS-OCTA showed that the CNV appeared to arise from an adjacent choroidal nevus. This is the first report of using SS-OCTA to diagnose CNV associated with a choroidal nevus masquerading as neovascular AMD. [Ophthalmic Surg Lasers Imaging Retina. 2018;49:360-363.].

Intravitreal bevacizumab associated with photodynamic therapy in a case of polypoidal choroidal vasculopathy associated with choroidal nevus: A case report

RATIONALE

Report the clinical findings and management of a case of polypoidal choroidal vasculopathy associated with choroidal nevus which received combination therapy.

PATIENT CONCERNS

Decreased visual acuity in a woman with polypoidal choroidal vasculopathy and choroidal nevus.

DIAGNOSES

Polypoidal choroidal vasculopathy and choroidal nevus.

INTERVENTIONS

The initial visual acuity was 0.5. After the first treatment with photodynamic therapy, exudation and bleeding appeared around the lesion. After this, the patient received 3 doses of intravitreal bevacizumab.

OUTCOMES

After treatment with combination therapy, visual acuity, clinical and imaging findings improved, with no recurrence of exudation and bleeding.

LESSONS

Intravitreal bevacizumab as an adjunctive treatment after photodynamic therapy is a good option for patients with polypoidal choroidal vasculopathy associated with choroidal nevus.

En Face Optical Coherence Tomography Angiography Imaging Versus Fundus Photography in the Measurement of Choroidal Nevi

BACKGROUND AND OBJECTIVES

Choroidal nevi are common benign intraocular tumors with a small risk of malignant transformation. This retrospective study investigates the use of en face spectral-domain optical coherence tomography angiography (SD-OCTA) in determining the clinical features and measurement of choroidal nevi.

PATIENTS AND METHODS

Patients with choroidal nevi were imaged with both OCTA and a fundus photography device. Greatest longitudinal dimension (GLD), perpendicular dimension (PD), and the GLD/PD ratio were assessed on each device. Inter-device variation and intra- and inter-rater reliability analyses were performed.

RESULTS

Fourteen patients with choroidal nevi were included. No significant difference between the GLD/PD ratio as measured by all three devices was found (Chi-square = 2.8, 2 df, P = .247). Intraclass correlation coefficients were greater than 0.7 for repeated measures on all devices, suggesting good repeatability and reproducibility.

CONCLUSION

This study demonstrated inter-device consistency and high intra- and inter-rater reliability when measuring choroidal nevi. [Ophthalmic Surg Lasers Imaging Retina. 2017;48:741-747.].

Complex Truncal Masses in the Setting of CLOVES Syndrome: Aesthetic and Functional Implications

BACKGROUND

Congenital lipomatous overgrowth, vascular malformations, epidermal nevi, and skeletal abnormalities (CLOVES) is a complex overgrowth syndrome with dramatic aesthetic and functional implications. The truncal masses characteristic of CLOVES syndrome are described as vascular malformations or lipomatous lesions with variable vascular components. Herein, we describe our single-institution experience with surgical excision of CLOVES-related truncal masses and discuss future directions in treatment of these complex anomalies.

METHODS

A single-institution retrospective review was performed for patients diagnosed with CLOVES syndrome. Patients undergoing excision of truncal vascular malformations were included. Outcome measures included perioperative characteristics [estimated blood loss (EBL), specimen size/anatomic location, blood-product requirement], as well as length-of-stay [LOS], and complication profile. Mean follow-up was 23.4 months (range 4.2-44).

RESULTS

Three consecutive patients were reviewed, accounting for 4 surgical operations. One patient underwent two operations for two distinct masses. All lesions were located on the upper back or flank with various degrees of muscular involvement. One patient required no transfusions with an uneventful 2-day hospitalization. The remaining three patients had an EBL ranging from 1500 to 6450 mL, requiring 9-13 units of packed red blood cells and 5-8 units of fresh frozen plasma during LOS (averaging 5 days). Mean weight of resected masses was 6.26 lbs (range 2.04-12 lbs) and mass dimensions ranged between 1778.9 and 15,680 cm³. One patient with recurrence was subsequently treated with a combination of sclerotherapy and rapamycin, leading to significant mass reduction.

CONCLUSIONS

Management of CLOVES syndrome requires a collaborative and multimodal approach. Although surgical debulking is one treatment option, non-invasive medical modalities and sclerotherapy should be considered prior to surgical resection.

LEVEL OF EVIDENCE IV

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Becker Naevus Syndrome of the Lower Body: One Case and Review of the Literature

Becker naevus syndrome is a rare epidermal naevus syndrome defined by the co-occurrence of a Becker naevus with various cutaneous, muscular and skeletal anomalies. In the majority of cases, abnormalities exclusively consist of ipsilateral hypoplasia of the breast, areola and/or nipple in addition to the naevus. Here, we report on a 42-year-old woman with an extensive Becker naevus reaching from the left buttock to the left calf verified on histological examination. In addition, there was marked hypoplasia of the fatty tissue of the left thigh confirmed by magnetic resonance imaging in contrast to hyperplasia of the fatty tissue of the left gluteal area. Underlying muscles and bones were not affected. There was no difference in leg lengths. In addition, we review and discuss the features of Becker naevus syndrome with emphasis on 10 reported cases with involvement of the lower body.

Handheld optical-resolution photoacoustic microscopy

Optical-resolution photoacoustic microscopy (OR-PAM) offers label-free Collapse

Key Words

• photoacoustic handheld probe
• optical-resolution photoacoustic microscopy
• two-axis microelectromechanical system mirror
• confocal scanning
• human skin imaging

Collapse

MESH Headings

• Animals
• Capillaries/diagnostic imaging
• Ear/blood supply
• Ear/diagnostic imaging
• Humans
• Mice
• Micro-Electrical-Mechanical Systems
• Microscopy, Acoustic
• Nevus/blood supply
• Nevus/diagnostic imaging
• Photoacoustic Techniques/instrumentation
• Spectrum Analysis

Collapse

Grants

• DP1 EB016986 NIBIB NIH HHS
• R01 CA159959 NCI NIH HHS
• R01 CA186567 NCI NIH HHS

Collapse

Affiliation(s)

Li Lin Washington University in St. Louis, Optical Imaging Laboratory, Department of Biomedical Engineering, One Brookings Drive, St. Louis, Missouri 63130, United States
Pengfei Zhang Washington University in St. Louis, Optical Imaging Laboratory, Department of Biomedical Engineering, One Brookings Drive, St. Louis, Missouri 63130, United States
Song Xu Texas A&M University, Institute for Solid State Electronics, Electrical Engineering Department, 400 Bizzell Street, College Station, Texas 77840, United States
Junhui Shi Washington University in St. Louis, Optical Imaging Laboratory, Department of Biomedical Engineering, One Brookings Drive, St. Louis, Missouri 63130, United States
Lei Li Washington University in St. Louis, Optical Imaging Laboratory, Department of Biomedical Engineering, One Brookings Drive, St. Louis, Missouri 63130, United States
Junjie Yao Washington University in St. Louis, Optical Imaging Laboratory, Department of Biomedical Engineering, One Brookings Drive, St. Louis, Missouri 63130, United States
Lidai Wang Washington University in St. Louis, Optical Imaging Laboratory, Department of Biomedical Engineering, One Brookings Drive, St. Louis, Missouri 63130, United States
Jun Zou Texas A&M University, Institute for Solid State Electronics, Electrical Engineering Department, 400 Bizzell Street, College Station, Texas 77840, United States
Lihong V. Wang Washington University in St. Louis, Optical Imaging Laboratory, Department of Biomedical Engineering, One Brookings Drive, St. Louis, Missouri 63130, United States

Collapse

Separating melanin from hemodynamics in nevi using multimode hyperspectral dermoscopy and spatial frequency domain spectroscopy

Changes in the pattern and distribution of both melanocytes (pigment producing) and vasculature (hemoglobin containing) are important in distinguishing melanocytic proliferations. The ability to accurately measure melanin distribution at different depths and to distinguish it from hemoglobin is clearly important when assessing pigmented lesions (benign versus malignant). We have developed a multimode hyperspectral dermoscope (SkinSpect™) able to more accurately image both melanin and hemoglobin distribution in skin. SkinSpect uses both hyperspectral and polarization-sensitive measurements. SkinSpect’s higher accuracy has been obtained by correcting for the effect of melanin absorption on hemoglobin absorption in measurements of melanocytic nevi. In vivo human skin pigmented nevi (N=20) were evaluated with the SkinSpect, and measured melanin and hemoglobin concentrations were compared with spatial frequency domain spectroscopy (SFDS) measurements. We confirm that both systems show low correlation of hemoglobin concentrations with regions containing different melanin concentrations (R=0.13 for SFDS, R=0.07 for SkinSpect).

Multi-scale classification based lesion segmentation for dermoscopic images

This paper presents a robust segmentation method based on multi-scale classification to identify the lesion boundary in dermoscopic images. Our proposed method leverages a collection of classifiers which are trained at various resolutions to categorize each pixel as "lesion" or "surrounding skin". In detection phase, trained classifiers are applied on new images. The classifier outputs are fused at pixel level to build probability maps which represent lesion saliency maps. In the next step, Otsu thresholding is applied to convert the saliency maps to binary masks, which determine the border of the lesions. We compared our proposed method with existing lesion segmentation methods proposed in the literature using two dermoscopy data sets (International Skin Imaging Collaboration and Pedro Hispano Hospital) which demonstrates the superiority of our method with Dice Coefficient of 0.91 and accuracy of 94%.

Melanoma detection by analysis of clinical images using convolutional neural network

Melanoma, most threatening type of skin cancer, is on the rise. In this paper an implementation of a deep-learning system on a computer server, equipped with graphic processing unit (GPU), is proposed for detection of melanoma lesions. Clinical (non-dermoscopic) images are used in the proposed system, which could assist a dermatologist in early diagnosis of this type of skin cancer. In the proposed system, input clinical images, which could contain illumination and noise effects, are preprocessed in order to reduce such artifacts. Afterward, the enhanced images are fed to a pre-trained convolutional neural network (CNN) which is a member of deep learning models. The CNN classifier, which is trained by large number of training samples, distinguishes between melanoma and benign cases. Experimental results show that the proposed method is superior in terms of diagnostic accuracy in comparison with the state-of-the-art methods.

Immunohistochemical expression of cyclooxygenase-2 (COX-2) in oral nevi and melanoma

BACKGROUND

Cyclooxygenase-2 (COX-2) catalyses the conversion of arachidonic acid to prostaglandin, and its overexpression has been demonstrated in different malignant tumors, including cutaneous melanoma. However, no data about the expression of this protein in oral melanocytic lesions are available to date. The aim of this study was to evaluate the immunohistochemical expression of COX-2 in oral nevi and melanomas, comparing the results with correspondent cutaneous lesions.

METHODS

COX-2 was evaluated by immunohistochemistry in 49 oral melanocytic lesions, including 36 intramucosal nevi and 13 primary oral melanomas, and in four cutaneous nevi and eight melanomas.

RESULTS

All cases of oral and cutaneous melanomas were positive for COX-2. On the other hand, all oral and cutaneous melanocytic nevi were negative.

CONCLUSION

COX-2 is highly positive in oral melanomas and negative in oral nevi and might represent a useful marker to distinguish melanocytic lesions of the oral cavity.

Nevus and melanoma paraconsistent classification

This paper presents the first studies on Nevus and Melanoma classification by using Paraconsistent Artificial Neural Network (PANN). Nevus is usually a small growth on the skin while Melanoma is a dangerous skin cancer. The proposed automated process classifies a set of medical images as Nevus and Melanoma based on a methodology grounded on PANN which is able to deal with conflicting, paracomplete and imprecise data directly without trivialization. Such methodology performed promising results considering only border features to classify the sample.

In vivo photoacoustic microscopy of human cutaneous microvasculature and a nevus

In several human volunteers, photoacoustic microscopy (PAM) has been utilized for noninvasive cutaneous imaging of the skin microvasculature and a melanocytic nevus. Microvascular networks in both acral and nonacral skin were imaged, and multiple features within the skin have been identified, including the stratum corneum, epidermal-dermal junction, and subpapillary vascular plexus. Several vascular and structural differences between acral and nonacral skin were also observed in the photoacoustic images. In addition, a nevus was photoacoustically imaged, excised, and histologically analyzed. The photoacoustic images allowed for in vivo measurement of tumor thickness, depth, and microvasculature-values confirmed by histologic examination. The presented images demonstrate the potential of PAM to aid in the study and evaluation of cutaneous microcirculation and analysis of pigmented lesions. Through its ability to three-dimensionally image the structure and function of the microvasculature and pigmented lesions, PAM can have a clinical impact in diagnosis and assessment of systemic diseases that affect the microvasculature such as diabetes and cardiovascular disease, cutaneous malignancies such as melanoma, and potentially other skin disorders.

The Use of Ultrasound Biomicroscopy in the Evaluation of Anterior Segment Tumors and Simulating Conditions

PURPOSE

To report the ultrasound biomicroscopy (UBM) findings of anterior segment tumors and simulating conditions.

METHODS

Thirty-five patients underwent UBM. Of those, 16 had histopathologically or cytopathologically diagnosed tumors, and 19 had clinically diagnosed lesions.

RESULTS

The study material comprised 13 iris pigment epithelial (IPE) cysts, 7 ciliary body melanomas, 4 iris melanomas, 4 iris nevi, 3 intraocular invasions of conjunctival squamous cell carcinoma, 2 ring melanomas of the anterior chamber angle, 1 medulloepithelioma and 1 pars plana cyst. On UBM, all IPE cysts presented as cystic lesions with a thin cyst wall and no solid components. All ciliary body melanomas showed low to medium reflectivity, with cavitation in one case and extraocular extension in another. Iris melanomas presented as anterior (stromal) iris lesions with medium to high internal reflectivity. There was irregularity and convex bowing of the posterior iris plane in iris melanomas, a feature not seen in iris nevi. Intraocular invasion of conjunctival squamous cell carcinoma was evidenced as areas of medium to high reflectivity in the ciliary body and iris, loss of the acute angle shape and highly reflective spots in the anterior chamber.

CONCLUSIONS

UBM was particularly useful in the diagnosis of IPE cysts, in the visualization of small ciliary body melanomas, in the differentiation of iris melanomas from iris nevi and in the demonstration of intraocular invasion from conjunctival squamous cell carcinoma.

Preoperative Ultrasonic Assessment of Thin Melanocytic Skin Lesions Using a 100-MHz Ultrasound Transducer: A Comparative Study

BACKGROUND

It has been shown that tumor thickness (TT) of melanocytic skin lesions (MSL) of less than 1 mm vertical thickness assessed by 20 MHz are often incorrectly evaluated.

OBJECTIVE

We aimed to evaluate the accuracy of 100-MHz ultrasound for the determination of TT of thin MSL, compared with conventional 20-MHz ultrasound and histologic findings.

METHODS

Thirty-seven patients with 50 suspicious MSL, including tumor diameter up to 1 cm and maximum vertical TT of less than 1 mm, were recruited. The agreement between the histologically and ultrasographically measured TT was analyzed using Bland and Altman plots.

RESULTS

Compared to histology, 20-MHz ultrasound (33.9 microm) as well as 100-MHz (16 microm) resulted in overestimation of TT that was twofold higher for 20-MHz ultrasound. The latter also revealed wider 95% limits of agreement (4.9 to 63 microm) than 100-MHz ultrasound (3.5 to 28.7 microm).

CONCLUSION

Analysis of agreement clearly demonstrated that the performance of 100-MHz ultrasound is superior to conventional 20-MHz ultrasound, even though a relatively small positive bias was observed in 100-MHz ultrasound, indicating a systematic error. We consider 100-MHz ultrasound a useful tool for the noninvasive determination of TT of thin MSL in vivo.

Quantitative discrimination of pigmented lesions using three-dimensional high-resolution ultrasound reflex transmission imaging

High-resolution ultrasound-reflex transmission imaging is a non-invasive method that can be performed in vivo. We have adapted and refined this technique for skin imaging. Scans can be analyzed to produce objective parameters. Previous work has highlighted sonographic differences between benign and malignant lesions. The aim of this study was to produce and test numerical parameters from ultrasound skin images that would quantify the acoustic differences between common pigmented lesions, which may aid their discrimination from melanoma. We report our findings for randomly selected patients referred from primary care with suspected melanoma. Those subsequently classified as malignant melanoma (MM), seborrheic keratosis (SK), and benign nevi by a consultant dermatologist (n=87) were imaged by high-resolution ultrasound-reflex transmission imaging. Using surrounding normal skin as a control, numerical sonographic parameters were derived for each lesion giving a relative measure of surface sound reflectance, intra-lesional sound reflection, total sound attenuation, and the relative uniformity of each parameter across the tumor. Significant quantitative differences existed between benign and malignant pigmented lesions studied. Sufficient discrimination was produced between MM (n=25), SKs (n=24) and other benign-pigmented lesions (n=38) to potentially reduce the referral of benign tumors by 65% without missing melanoma.

Long term ultrasonic follow up of choroidal naevi and their transformation to melanomas

AIMS

To compare ultrasonographic (US) predicting factors for conversion of choroidal naevi into melanomas.

METHODS

659 consecutive eyes with choroidal naevi were examined between 1984 and 2004. 165 clinically suspicious naevi were followed clinically and ultrasonographically (thickness, base diameters, internal reflectivity and location in the eye) for 5.08 (SE 0.24) years.

RESULTS

17 naevi (2.6% of all naevi, 10.3% of suspicious naevi) converted to small choroidal melanomas. The thickness of benign and premalignant naevi differed significant only after 1.5 years of follow up. The mean initial thickness of benign and premalignant naevi was significantly different (p = 0.001), as was mean initial internal reflectivity (p = 0.002) and mean initial largest base diameter (LBD, p = 0.05). Posterior pole and nasally located naevi were more likely to become malignant. A thickness of > or = 2 mm and a LBD > or = 7 mm were most predictive of conversion to melanoma, as was a combined K(I) index of > or = 14.5 (KI = LBD + 4 x thickness + 1 (for nasal location) + 1 (for posterior pole location)). An artificial neural network did not have a better forecasting accuracy than the KI index. Logistic regression found the only significant parameters to influence the risk of conversion to melanoma to be the KI value and the initial tumour thickness.

CONCLUSIONS

A follow up of at least 1.5 years is necessary to detect conversion of naevi to choroidal melanomas. The thickness and LBD of the lesion can be used for predicting the risk.

Real-time spatial compound ultrasound imaging of skin

PURPOSE

The aim of our study was to evaluate the potential of real-time spatial compound imaging (RTSCI) in dermatology.

MATERIALS AND METHODS

An ATL 5000 SonoCT equipped with compact linear 15-7 MHz and linear 12-5 MHz transducers was obtained for skin visualization in a group of dermatological patients with various skin diseases.

RESULTS

Thirty-four people participated: 21 patients with various skin diseases and 13 persons with normal skin. The mean age was 43.4 years. For many diseases, RTSCI gave useful information about the lesional structure, thickness and relationship with surrounding structures.

CONCLUSION

RTSCI allows objective, accurate, noninvasive and easy measurements of several parameters of skin morphology. It is useful in clinical trials, for evaluation of the effects of therapy, for preoperative evaluation of dermatological lesions, and enables visualization of subclinical and deep lesions, giving physicians the possibility of starting treatment before disease intensity increases. However, even such highly advanced ultrasound cannot completely substitute the clinical dermatological approach and the occasional need for histological diagnosis. This new method may, however, become an important adjunct method for the study of skin lesions.

Needle-guided shave excision with ultrasonographic assistance: a new technique

BACKGROUND

A lack of control in the deep margin of shave excision is a drawback of this technique.

OBJECTIVE

To describe a more precise shaving technique by sonography with the use of a fine injector needle as a depth marker.

METHODS

After having examined the invasion levels of 40 benign skin lesions, a fine injector needle was intentionally threaded into the dermis horizontally just beneath the required shaving plane. Correct placement of the needle was ensured by sonographic examination after possible reinsertion trials. Large lesions needed several needles to be inserted. The tissue above the needles was then shaved off, whereas the deep dermal layer was protected.

RESULTS

Histologic examination revealed that 77% of superficial and intermediate-thickness lesions were removed totally without disturbing the derma-fat junction.

CONCLUSION

The needle-guided technique was found to be effective in the control of the deep margin of shave excision.

Linear sebaceous naevus syndrome and resistant rickets

The association between vitamin-D-resistant rickets and linear sebaceous naevus syndrome is extremely rare. Only eight cases have been described in the English literature and in none were the skeletal aspects addressed. We present three new cases and describe the musculoskeletal features. The details and outcome of surgery for correction of the deformities are discussed. The disturbances of metabolism of vitamin D and the effects of pharmacological treatment are also described.

Efficacy of (123)N-isopropyl-p-[(123)I]-iodoamphetamine single photon emission computed tomography for the diagnosis of uveal malignant melanoma

PURPOSE

To evaluate the efficacy of single photon emission computed tomography (SPECT) images with N-isopropyl-p-[(123)I]-iodoamphetamine ([(123)I]-IMP) as a radiopharmaceutical for the diagnosis of uveal melanoma.

METHODS

Consecutive interventional case series. Single photon emission computed tomography images were obtained after intravenous injection of [(123)I]-IMP in 20 patients with clinically suspected uveal melanoma (16 eyes of 16 patients) or ocular adnexal melanoma (four adnexal tumors of four patients). Histologic findings in seven eyes enucleated for uveal melanoma were analyzed with the SPECT images semiquantitatively evaluated.

RESULTS

Eight of 20 eyes with suspected uveal melanoma showed high accumulation of [(123)I]-IMP in the late phase in the area corresponding to the uveal tumor, and the diagnosis of uveal melanoma was confirmed histologically in seven enucleated eyes. Tumor size and histologic features did not reflect the intensity of scintillation counts of [(123)I]-IMP. Twelve patients with no accumulation on the SPECT images were found to have clinically or histologically iris nevus, choroidal nevus, and other intraocular and adnexal conditions.

CONCLUSION

N-isopropyl p-[(123)I]-iodoamphetamine SPECT may serve as a sensitive and specific examination for the diagnosis of uveal melanoma.

Classification of reflectance spectra from pigmented skin lesions, a comparison of multivariate discriminant analysis and artificial neural networks

Successful treatment of skin cancer, especially melanoma, depends on early detection, but diagnostic accuracy, even by experts, can be as low as 56% so there is an urgent need for a simple, accurate, non-invasive diagnostic tool. In this paper we have compared the performance of an artificial neural network (ANN) and multivariate discriminant analysis (MDA) for the classification of optical reflectance spectra (320 to 1100 nm) from malignant melanoma and benign naevi. The ANN was significantly better than MDA, especially when a larger data set was used, where the classification accuracy was 86.7% for ANN and 72.0% for MDA (p < 0.001). ANN was better at learning new cases than MDA for this particular classification task. This study has confirmed that the convenience of ANNs could lead to the medical community and patients benefiting from the improved diagnostic performance which can be achieved by objective measurement of pigmented skin lesions using spectrophotometry.

Evaluation of ocular tumors with technetium-99m-MIBI: planar pinhole technique or SPECT?

OBJECTIVE

This study compares 2 imaging protocols, planar pinhole technique (PPHT) and SPECT, for evaluating ocular masses with 99mTc-MIBI.

METHODS

Sixteen patients with ocular lesions were studied. Planar images were acquired 10 min after the injection of 740 MBq 99mTc-MIBI with an LFOV camera fitted with a pinhole collimator (5.0 mm). A SPECT study was performed immediately after the planar study, using a 360 degrees orbit, 64 steps, 20 s/stop, a 128 x 128 matrix, and a low-energy high-resolution (LEHR) collimator. Twelve lesions (9.5-18.0 mm) proved to be malignant: 8 primary tumors (ocular melanoma); 3 local relapses of different tumors of the conjunctiva; and 1 ocular metastasis from breast cancer. The remaining 4 lesions (10.0-16.0 mm) were benign: 1 inflammatory lesion; 1 benign intraocular calcification; and 2 naevi.

RESULTS

SPECT images showed 11 of 12 malignant lesions (91.6%), whereas the planar technique demonstrated only 4 of the 12 lesions (33.3%). One false-positive result, the inflammatory lesion, was visualized by both techniques. The remaining benign lesions were not detected with either method.

CONCLUSION

Technetium-99m-MIBI SPECT is a sensitive technique for detecting malignant ocular tumors. SPECT imaging is a better alternative to planar imaging for ocular tumors.

Ultrasound biomicroscopy. High-frequency ultrasound imaging of the eye at microscopic resolution

UBM presents us with a new method of imaging the anterior segment of the eye at high resolution. Its strengths lie in its ability to produce cross-sections of the living eye at microscopic resolution without violating the integrity of the globe. UBM, although lacking the resolution of optical microscopy, gives us images in living eyes without affecting the internal relationships of the structures imaged. There are many other applications of this new imaging method. Examples of other uses include imaging adnexal pathology, assessing corneal changes with refractive surgery, the assessment of trauma, and determination of intraocular lens position.

Skeletal involvement and follow-up in linear nevus sebaceous syndrome

Linear nevus sebaceous syndrome is a rare neurocutaneous syndrome in which associated radiographic findings of the peripheral skeletal system are uncommon. We report a follow-up study of a patient with mainly unilateral changes of the skeletal system involving the thorax and the right arm and leg.

[The primary study of ultrasound biomicroscope in imaging anterior segment tumors of eye]

PURPOSE

To evaluate the use of ultrasound biomicroscope (UBM) in imaging anterior segment tumors of eye.

METHODS

13 cases of anterior segment tumors of eye were examined with UBM.

RESULTS

3 cases had the ocular history of trauma or surgery and were clinically diagnosed as anterior iris cysts. UBM displayed the cysts as thin-walled, no internal reflectivity, and at the locate of anterior humour angle; 8 cases had superficial anterior humour, narrow anterior angle and clinically be suspected as tumors behind iris or ciliary body. UBM showed cystic reflectivity in 7 cases and a solid tumor in ciliary body in one case. 2 cases were clinically diagnosed as iris nevi. UBM displayed solid convex lesions.

CONCLUSION

UBM is a valuable new technique in the diagnosis and accurate localization of anterior segment tumors.

Value of high-frequency US for preoperative assessment of skin tumors

The purpose of this study was to evaluate the accuracy of high-frequency ultrasound (US) in the preoperative assessment of skin tumors. A US scanner with a 20-MHz probe was used to visualize and evaluate 70 skin lesions (38 clinically suspected melanomas and 32 suspected basilar cell carcinomas [BCCs]) before surgical resection. A US morphologic study and a Doppler analysis of vascularity were performed for each tumor. Of the 70 tumors, 62 were clearly visualized, including 19 melanomas, 12 nonmalignant nevi, and 31 BCCs. Most lesions were hypoechoic. In 13 of 19 proved melanomas, the difference between the histologic and US measurements was equal to or less than 0.2 mm. Vessels were visualized in melanomas with thicknesses greater than 3 mm. All BCCs were visualized, and in 29% of cases of BCC, tumor size at US was greater than that at clinical examination. High-frequency, high-resolution US is a simple, reliable, noninvasive method for accurate preoperative assessment of skin tumor dimensions. This technique allows surgical planning to be adapted and reexcision to be avoided. However, its role is limited in the differential diagnosis of malignant and benign skin lesions.

Visualizing ocular melanoma using iodine-123-N-(2-diethylaminoethyl)4-iodobenzamide SPECT

Radiolabeled benzamides have recently been introduced for the detection of melanoma. We evaluated the potential clinical applicability of 123I-N-(2-diethylaminoethyl) 4-iodobenzamide ([123I]IDAB) for SPECT imaging of ocular melanoma.

METHODS

Fourteen patients were studied, 10 with or suspected of malignant ocular melanoma and four with ocular naevi. All patients underwent SPECT imaging of the head and whole-body scintigraphy 4-5 hr after injection of 170 MBq [123I]IDAB.

RESULTS

A definite tracer hyperfixation was observed in the pathological eye in 9 of 10 (90%) patients with ocular melanoma. The pathological-to-normal eye ratio averaged 1.46 (range 1.07-2.86). The melanoma nature of the scintigraphic lesions was confirmed after enucleation in eight cases and by clinical evolution in two. A false-negative scan was reported in a patient with a small and hypochromic lesion. In patients with ocular naevi, no false-positive scintigrams were documented.

CONCLUSION

Iodine-123-IDAB scintigraphy may contribute significantly to decide about enucleation in cases where some doubt persists with conventional techniques.

[Objectivity, reproducibility and validity of 3D ultrasound in dermatology]

Recently it has become possible to visualize the three-dimensional (3D) architecture of sonographic structures using high-resolution ultrasound above 20 MHz (Dermascan C, Denmark; DUB 20, Germany). We studied the objectivity, reproducibility, and validity of these two 3D analysis systems on 22 skin tumors. To validate both systems, we compared the sonometric and histometric volume in 4 tumors. The objectivity was calculated by repeated evaluation of the volume on the same serial B-scan sections at 3-month intervals. The two-dimensional B-scans and 3D images exhibited similar characteristics. Volumetric evaluations were reproducible with both systems (variation coefficient < or = 9.9%), when the applicator was oriented longitudinal to the body axis. By changing the orientation, the coefficient increased up to 58.1% (DUB 20). Sonometric and histometric volumes of a phantom model and of the tumors exhibited small deviations in the Dermascan C (< or = 8.1%) and in some tumors large deviations in the DUB 20 (< or = 93.1%). We can conclude that the 3D sonography is a reproducible method, which can be used to evaluate the volume of skin tumors and metastases under chemotherapy or radiotherapy. We suggest in particular to use the in-vivo tumor volume as a prognostic parameter for the malignant melanoma.

Does high-frequency (40-60 MHz) ultrasound imaging play a role in the clinical management of cutaneous melanoma?

The assessment of cutaneous melanoma in the clinical setting is often difficult, and important features such as depth and width remain unknown until the pathology report is received. Access to prognostic features such as vertical height before excisional biopsy would offer a basis for guidance in defining surgical margins and early planning of treatment options. Recently developed high-frequency ultrasound imaging in the 40-to 60-MHz range is a noninvasive method that provides in vivo information about cutaneous lesions. Imaging at these frequencies provides high-resolution data within the range of the epidermis and dermis (3-4 mm in depth). Ten cutaneous melanomas and seven pigmented lesions were assessed in this fashion. Vertical height was documented and compared to histopathological findings. High-frequency ultrasound imaging determination of vertical height correlated well with the standard measurement of Breslow's thickness on histological sections only in midrange (1.0-3.0 mm) lesions. Inflammatory cells at the base of three melanomas provoked an overestimation of the depth measurement with ultrasonography. Thick keratin layers such as those found on the feet acted as a virtual block to the high-frequency scanner. The application of this new advance in noninvasive imaging technology to the clinical assessment of cutaneous melanoma provides interesting in vivo data but in its present state does not replace the need for the biopsy of pigmented lesions and histopathological diagnosis.

Epidermal nevus syndrome with isolated enlargement of one temporal lobe: a case report

A 5-month-old girl presented with infantile spasms and psychomotor retardation. She was born with epidermal nevi on the right side of the head, the right arm and the back. MRI of the head showed a right hemimegalencephaly limited to the temporal lobe. There were no MRI signs of neocortical dysplasia. Ultrasonography revealed multiple cysts in both ovaria and a small solid lesion in the left adrenal gland. Hemimegalencephaly represents an overgrowth phenomenon; the isolated temporal lobe hypertrophy can be explained as a mild expression of hemimegalencephaly. To our knowledge this has not been described before.