In this work, we propose to use an artificial neural network to classify limited data of clinical multispectral and autofluorescence images of skin lesions. Although the amount of data is limited, the deep convolutional neural network classification of skin lesions using a multi-modal image set is studied and proposed for the first time. The unique dataset consists of spectral reflectance images acquired under 526 nm, 663 nm, 964 nm, and autofluorescence images under 405 nm LED excitation. The augmentation algorithm was applied for multi-modal clinical images of different skin lesion groups to expand the training datasets. It was concluded from saliency maps that the classification performed by the convolutional neural network is based on the distribution of the major skin chromophores and endogenous fluorophores. The resulting classification confusion matrices, as well as the performance of trained neural networks, have been investigated and discussed.

U-Net is the most cited and widely-used deep learning model for biomedical image segmentation. In this paper, we propose a new enhanced version of a ubiquitous U-Net architecture, which improves upon the original one in terms of generalization capabilities, while addressing several immanent shortcomings, such as constrained resolution and non-resilient receptive fields of the main pathway. Our novel multi-path architecture introduces a notion of an individual receptive field pathway, which is merged with other pathways at the bottom-most layer by concatenation and subsequent application of Layer Normalization and Spatial Dropout, which can improve generalization performance for small datasets. In general, our experiments show that the proposed multi-path architecture outperforms other state-of-the-art approaches that embark on similar ideas of pyramid structures, skip-connections, and encoder–decoder pathways. A significant improvement of the Dice similarity coefficient is attained at our proprietary colony-forming unit dataset, where a score of 0.809 was achieved for the foreground class.

Melanoma is a melanocytic tumor that is responsible for the most skin cancer-related deaths. By contrast, seborrheic keratosis (SK) is a very common benign lesion with a clinical picture that may resemble melanoma. We used a multispectral imaging device to distinguish these two entities, with the use of autofluorescence imaging with 405 nm and diffuse reflectance imaging with 525 and 660 narrow-band LED illumination. We analyzed intensity descriptors of the acquired images. These included ratios of intensity values of different channels, standard deviation and minimum/maximum values of intensity of the lesions. The pattern of the lesions was also assessed with the use of particle analysis. We found significantly higher intensity values in SKs compared with melanoma, especially with the use of the autofluorescence channel. Moreover, we found a significantly higher number of particles with high fluorescence in SKs. We created a parameter, the SK index, using these values to differentiate melanoma from SK with a sensitivity of 91.9% and specificity of 57.0%. In conclusion, this imaging technique is potentially applicable to distinguish melanoma from SK based on the analysis of various quantitative parameters. For this application, multispectral imaging could be used as a screening tool by general physicians and non-experts in the everyday practice.

Colony-Forming Unit (CFU) counting is a complex problem without a universal solution in biomedical and food safety domains. A multitude of sophisticated heuristics and segmentation-driven approaches have been proposed by researchers. However, U-Net remains the most frequently cited and used deep learning method in these domains. The latter approach provides a segmentation output map and requires an additional counting procedure to calculate unique segmented regions and detect microbial colonies. However, due to pixel-based targets, it tends to generate irrelevant artifacts or errant pixels, leading to inaccurate and mixed post-processing results. In response to these challenges, this paper proposes a novel hybrid counting approach, incorporating a multi-loss U-Net reformulation and a post-processing Petri dish localization algorithm. Firstly, a unique innovation lies in the multi-loss U-Net reformulation. An additional loss term is introduced in the bottleneck U-Net layer, focusing on the delivery of an auxiliary signal that indicates where to look for distinct CFUs. Secondly, the novel localization algorithm automatically incorporates an agar plate and its bezel into the CFU counting techniques. Finally, the proposition is further enhanced by the integration of a fully automated solution, which comprises a specially designed uniform Petri dish illumination system and a counting web application. The latter application directly receives images from the camera, processes them, and sends the segmentation results to the user. This feature provides an opportunity to correct the CFU counts, offering a feedback loop that contributes to the continued development of the deep learning model. Through extensive experimentation, the authors of this paper have found that all probed multi-loss U-Net architectures incorporated into the proposed hybrid approach consistently outperformed their single-loss counterparts, as well as other comparable models such as self-normalized density maps and YOLOv6, by at least 1% to 3% in mean absolute and symmetric mean absolute percentage errors. Further significant improvements were also reported through the means of the novel localization algorithm. This reaffirms the effectiveness of the proposed hybrid solution in addressing contemporary challenges of precise in vitro CFU counting.

Mobile platform for multimodal skin assessment has been developed. Different illumination sources allow switching between modalities. Diffuse reflectance spectral imaging is provided by five LEDs, fluorescence is excited by 405 nm LEDs, and laser speckle by 650 nm laser diode. Handheld, battery powered device includes all light sources and color camera with USB. The core of the system is Linux OS embedded microcontroller. USB, FTP and HTML with JavaScript combination is used to create standard image transfer and control interface. In combination with built in WiFi access point it allows online skin images storage and processing. It means that data processing algorithms are located and updated online.

Background/Objectives: The timely diagnostics of bladder cancer is still a challenge in clinical settings. The reliability of conventional testing methods does not reach desirable accuracy and sensitivity, and it has an invasive nature. The present study examines the application of machine learning to improve bladder cancer diagnostics by integrating miRNA expression levels, demographic routine laboratory test results, and clinical data. We proposed that merging these datasets would enhance diagnostic accuracy. Methods: This study combined molecular biology methods for liquid biopsy, routine clinical data, and application of machine learning approach for the acquired data analysis. We evaluated urinary exosome miRNA expression data in combination with patient test results, as well as clinical and demographic data using three machine learning models: Random Forest, SVM, and XGBoost classifiers. Results: Based solely on miRNA data, the SVM model achieved an ROC curve area of 0.75. Patient analysis’ clinical and demographic data obtained ROC curve area of 0.80. Combining both data types enhanced performance, resulting in an F1 score of 0.79 and an ROC of 0.85. The feature importance analysis identified key predictors, including erythrocytes in urine, age, and several miRNAs. Conclusions: Our findings indicate the potential of a multi-modal approach to improve the accuracy of bladder cancer diagnosis in a non-invasive manner.