Real-time supervised detection of pink areas in dermoscopic images of melanoma: importance of color shades, texture and location

BACKGROUND/PURPOSE

Early detection of malignant melanoma is an important public health challenge. In the USA, dermatologists are seeing more melanomas at an early stage, before classic melanoma features have become apparent. Pink color is a feature of these early melanomas. If rapid and accurate automatic detection of pink color in these melanomas could be accomplished, there could be significant public health benefits.

METHODS

Detection of three shades of pink (light pink, dark pink, and orange pink) was accomplished using color analysis techniques in five color planes (red, green, blue, hue, and saturation). Color shade analysis was performed using a logistic regression model trained with an image set of 60 dermoscopic images of melanoma that contained pink areas. Detected pink shade areas were further analyzed with regard to the location within the lesion, average color parameters over the detected areas, and histogram texture features.

RESULTS

Logistic regression analysis of a separate set of 128 melanomas and 128 benign images resulted in up to 87.9% accuracy in discriminating melanoma from benign lesions measured using area under the receiver operating characteristic curve. The accuracy in this model decreased when parameters for individual shades, texture, or shade location within the lesion were omitted.

CONCLUSION

Texture, color, and lesion location analysis applied to multiple shades of pink can assist in melanoma detection. When any of these three details: color location, shade analysis, or texture analysis were omitted from the model, accuracy in separating melanoma from benign lesions was lowered. Separation of colors into shades and further details that enhance the characterization of these color shades are needed for optimal discrimination of melanoma from benign lesions.

Discrimination of squamous cell carcinoma in situ from seborrheic keratosis by color analysis techniques requires information from scale, scale-crust and surrounding areas in dermoscopy images

Scale-crust, also termed "keratin crust", appears as yellowish-to-tan scale on the skin's surface. It is caused by hyperkeratosis and parakeratosis in inflamed areas of squamous cell carcinoma in situ (SCCIS, Bowen's disease) and is a critical dermoscopy feature for detecting this skin cancer. In contrast, scale appears as a white-to-ivory detaching layer of the skin, without crust, and is most commonly seen in benign lesions such as seborrheic keratoses (SK). Distinguishing scale-crust from ordinary scale in digital dermoscopy images holds promise for early SCCIS detection and differentiation from SK. Reported here are image analysis techniques that best characterize scale-crust in SCCIS and scale in SK, thereby allowing accurate separation of these two dermoscopic features. Classification using a logistic regression operating on color features extracted from these digital dermoscopy structures can reliably separate SCCIS from SK.

A deep learning approach to detect blood vessels in basal cell carcinoma

PURPOSE

Blood vessels called telangiectasia are visible in skin lesions with the aid of dermoscopy. Telangiectasia are a pivotal identifying feature of basal cell carcinoma. These vessels appear thready, serpiginous, and may also appear arborizing, that is, wide vessels branch into successively thinner vessels. Due to these intricacies, their detection is not an easy task, neither with manual annotation nor with computerized techniques. In this study, we automate the segmentation of telangiectasia in dermoscopic images with a deep learning U-Net approach.

METHODS

We apply a combination of image processing techniques and a deep learning-based U-Net approach to detect telangiectasia in digital basal cell carcinoma skin cancer images. We compare loss functions and optimize the performance by using a combination loss function to manage class imbalance of skin versus vessel pixels.

RESULTS

We establish a baseline method for pixel-based telangiectasia detection in skin cancer lesion images. An analysis and comparison for human observer variability in annotation is also presented.

CONCLUSION

Our approach yields Jaccard score within the variation of human observers as it addresses a new aspect of the rapidly evolving field of deep learning: automatic identification of cancer-specific structures. Further application of DL techniques to detect dermoscopic structures and handle noisy labels is warranted.

Heads or tails? Differential translational regulation in cercarial heads and tails of schistosome worms

Schistosomes are obligate helminths responsible for over 218 million cases of human schistosomiasis in 78 countries around the world. Infection occurs when free-swimming cercariae penetrate human skin and initiate developmental progression into parasitic obligate worms that consume red blood cells. Transcriptomic studies of infectious cercariae reveal abundant mRNAs associated with energy metabolism and host invasion. However, the cercaria is mostly transcriptionally quiescent, suggesting that most mRNAs are primed prior to cercarial escape from the snail host. The use of transcriptomics to understand protein expression presumes that transcription and translation are functionally coupled and the cercarial stage has categorically been treated as a single unit for -omic analysis. Per contra, the relationship between transcription and translation in infectious cercariae has not been described. To understand the correlation between transcription and translation in cercariae, we separately measured nascent translation levels in cercarial heads, cercarial tails and in the developing schistosomula, the next stage of its life cycle. The loss of the cercarial tail is essential for the transformation from a cercaria to a schistosomulum. We observed that translation was initially limited and the translation rate accelerated during the first 72-hours after tail loss. When we tested nascent translation in cercarial heads, cercarial tails, whole cercariae, and 4-hour schistosomula, we found that translation is significantly upregulated in the cercarial tail when compared to the cercarial head and that translation was undetectable in heads using immunofluorescent image quantification (p = .0005). These data represent a major shift in how we understand the cercarial stage. The cercarial head is mostly transcriptionally and translationally quiescent while being sufficient for progression into a schistosomulum. In addition, transcription and translation are not linked in Schistosoma mansoni cercaria. Thus, our current conceptual approach of treating the cercaria as a single functional unit for -omic studies may be insufficient to understand cercarial development.

Management of the very weak pelvic floor. Is there a point?

Conservative treatment in the form of pelvic muscle exercises is effective in the treatment of stress urinary incontinence. There are no studies specifically looking at women who have extremely weak pelvic muscles and their response to conservative treatment. This study looks at the effectiveness of pelvic muscle therapy in women with very weak pelvic muscles. Out of 965 women attending the bladder clinic at Kirwan Hospital, 219 were assessed to have weak pelvic muscles by digital palpation. All patients were subjected to a detailed urogynecological questionnaire, a frequency/volume chart, and clinical assessment. All patients were then given detailed verbal and written instructions on good bladder habits, including posture, dietetic habits and pelvic muscle exercises. Of the 219 women, 163 were able to complete their treatment and presented for review. Out of 163 women 118 (72%) reported a subjective improvement; 89 (54%) demonstrated an objective improvement in pelvic muscle strength.

Multiomic analysis of Schistosoma mansoni reveals unique expression profiles in cercarial heads and tails

Schistosomes require both molluscan and mammalian hosts for development. The larval cercaria exits the snail host and swims to identify and invade the mammalian host. The cercaria has two macrostructures, the head and the tail. The head invades the host, where it matures into an adult worm. The tail is lost after host invasion. Translation in the cercaria differs in each macrostructure, with higher levels of translation in the cercarial tail and little to no translational activity in the cercarial head. We compared the transcriptome and proteome of the cercarial head and tail and observed stark differences between the two macrostructures. We identified unique and differentially expressed transcripts and proteins, including ribosomal components expressed in higher levels in tails than in heads, which may explain the differences in translation levels between heads and tails. We also characterized the weak correlation between transcription and translation in infectious cercarial heads and tails.

Characterization of Schistosome Sox Genes and Identification of a Flatworm Class of Sox Regulators

Schistosome helminths infect over 200 million people across 78 countries and are responsible for nearly 300,000 deaths annually. However, our understanding of basic genetic pathways crucial for schistosome development is limited. The sex determining region Y-box 2 (Sox2) protein is a Sox B type transcriptional activator that is expressed prior to blastulation in mammals and is necessary for embryogenesis. Sox expression is associated with pluripotency and stem cells, neuronal differentiation, gut development, and cancer. Schistosomes express a Sox-like gene expressed in the schistosomula after infecting a mammalian host when schistosomes have about 900 cells. Here, we characterized and named this Sox-like gene SmSOXS1. SmSoxS1 protein is a developmentally regulated activator that localizes to the anterior and posterior ends of the schistosomula and binds to Sox-specific DNA elements. In addition to SmSoxS1, we have also identified an additional six Sox genes in schistosomes, two Sox B, one SoxC, and three Sox genes that may establish a flatworm-specific class of Sox genes with planarians. These data identify novel Sox genes in schistosomes to expand the potential functional roles for Sox2 and may provide interesting insights into early multicellular development of flatworms.