A deep learning approach for Direct Immunofluorescence pattern recognition of Autoimmune Bullous Diseases.

BACKGROUND: Artificial intelligence (AI) is reshaping healthcare, using machine and deep learning to enhance disease management. Dermatology has seen improved diagnostics, particularly in skin cancer detection, through the integration of AI. However, the potential of AI in automating immunofluorescence imaging for autoimmune bullous skin diseases remains untapped. While direct immunofluorescence (DIF) supports diagnosis, its manual interpretation can hinder efficiency. The use of deep learning to automatically classify DIF patterns, including the Intercellular Pattern (ICP) and the Linear Pattern (LP), holds promise for improving the diagnosis of autoimmune bullous skin diseases. OBJECTIVES: The objectives of this study are to develop AI algorithms for automated classification of autoimmune bullous skin disease DIF patterns, such as ICP and LP. This aims to enhance diagnostic accuracy, streamline disease management, and improve patient outcomes through deep learning-driven immunofluorescence interpretation. METHODS: We collected immunofluorescence images from skin biopsies of patients suspected of AIBD between January 2022 and January 2024. Skin tissue was obtained via 5-mm punch biopsy, prepared for direct immunofluorescence. Experienced dermatologists classified the images into three classes: ICP, LP, and negative. To evaluate our deep learning approach, we divided the images into training (436) and test sets (93). We employed transfer learning with pre-trained deep neural networks and conducted 5-fold cross-validation to assess model performance. Our dataset's class imbalance was addressed using weighted loss and data augmentation strategies. The models were trained for 50 epochs using Pytorch, achieving an image size of 224x224 for both CNNs and the Swin Transformer. RESULTS: Our study compared six CNNs and the Swin transformer for AIBDs image classification, with the Swin transformer achieving the highest average validation accuracy of 98.5%. On a separate test set, the best model attained an accuracy of 94.6%, demonstrating 95.3% sensitivity and 97.5% specificity across AIBDs classes. Visualization with Grad-CAM highlighted the model's reliance on characteristic patterns for accurate classification. CONCLUSIONS: The study highlighted CNN's accuracy in identifying DIF features. This approach aids automated analysis and reporting, offering reproducibility, speed, data handling, and cost-efficiency. Integrating deep learning in skin immunofluorescence promises precise diagnostics and streamlined reporting in this branch of dermatology.

An efficient deep learning approach to identify dynamics in in vitro neural networks.

Understanding and discriminating the spatiotemporal patterns of activity generated by in vitro and in vivo neuronal networks is a fundamental task in neuroscience and neuroengineering. The state-of-the-art algorithms to describe the neuronal activity mostly rely on global and local well-established spike and burst-related parameters. However, they are not able to capture slight differences in the activity patterns. In this work, we introduce a deep-learning-based algorithm to automatically infer the dynamics exhibited by different neuronal populations. Specifically, we demonstrate that our algorithm is able to discriminate with high accuracy the dynamics of five different populations of in vitro human-derived neural networks with an increasing inhibitory to excitatory neurons ratio.

Deepening the role of excitation/inhibition balance in human iPSCs-derived neuronal networks coupled to MEAs during long-term development.

Objective.The purpose of this study is to investigate whether and how the balance between excitation and inhibition ('E/I balance') influences the spontaneous development of human-derived neuronal networksin vitro. To achieve that goal, we performed a long-term (98 d) characterization of both homogeneous (only excitatory or inhibitory neurons) and heterogeneous (mixed neuronal types) cultures with controlled E/I ratios (i.e. E:I 0:100, 25:75, 50:50, 75:25, 100:0) by recording their electrophysiological activity using micro-electrode arrays.Approach.Excitatory and inhibitory neurons were derived from human induced pluripotent stem cells (hiPSCs). We realized five different configurations by systematically varying the glutamatergic and GABAergic percentages.Main results.We successfully built both homogeneous and heterogeneous neuronal cultures from hiPSCs finely controlling the E/I ratios; we were able to maintain them for up to 3 months. Homogeneity differentially impacted purely inhibitory (no bursts) and purely excitatory (few bursts) networks, deviating from the typical traits of heterogeneous cultures (burst dominated). Increased inhibition in heterogeneous cultures strongly affected the duration and organization of bursting and network bursting activity. Spike-based functional connectivity and image-based deep learning analysis further confirmed all the above.Significance.Healthy neuronal activity is controlled by a well-defined E/I balance whose alteration could lead to the onset of neurodevelopmental disorders like schizophrenia or epilepsy. Most of the commonly usedin vitromodels are animal-derived or too simplified and thus far from thein vivohuman condition. In this work, by performing a long-term study of hiPSCs-derived neuronal networks obtained from healthy human subjects, we demonstrated the feasibility of a robustin vitromodel which can be further exploited for investigating pathological conditions where the E/I balance is impaired.

In-domain versus out-of-domain transfer learning in plankton image classification.

Plankton microorganisms play a huge role in the aquatic food web. Recently, it has been proposed to use plankton as a biosensor, since they can react to even minimal perturbations of the aquatic environment with specific physiological changes, which may lead to alterations in morphology and behavior. Nowadays, the development of high-resolution in-situ automatic acquisition systems allows the research community to obtain a large amount of plankton image data. Fundamental examples are the ZooScan and Woods Hole Oceanographic Institution (WHOI) datasets, comprising up to millions of plankton images. However, obtaining unbiased annotations is expensive both in terms of time and resources, and in-situ acquired datasets generally suffer from severe imbalance, with only a few images available for several species. Transfer learning is a popular solution to these challenges, with ImageNet1K being the most-used source dataset for pre-training. On the other hand, datasets like the ZooScan and the WHOI may represent a valuable opportunity to compare out-of-domain and large-scale plankton in-domain source datasets, in terms of performance for the task at hand.In this paper, we design three transfer learning pipelines for plankton image classification, with the aim of comparing in-domain and out-of-domain transfer learning on three popular benchmark plankton datasets. The general framework consists in fine-tuning a pre-trained model on a plankton target dataset. In the first pipeline, the model is pre-trained from scratch on a large-scale plankton dataset, in the second, it is pre-trained on large-scale natural image datasets (ImageNet1K or ImageNet22K), while in the third, a two-stage fine-tuning is implemented (ImageNet [Formula: see text] large-scale plankton dataset [Formula: see text] target plankton dataset). Our results show that an out-of-domain ImageNet22K pre-training outperforms the plankton in-domain ones, with an average boost in test accuracy of around 6%. In the next part of this work, we adopt three ImageNet22k pre-trained Vision Transformers and one ConvNeXt, obtaining results on par (or slightly superior) with the state-of-the-art, corresponding to the usage of CNN models ensembles, with a single model. Finally, we design and test an ensemble of our Vision Transformers and the ConvNeXt, outperforming the state-of-the-art existing works on plankton image classification on the three target datasets. To support scientific community contribution and further research, our implemented code is open-source and available at https://github.com/Malga-Vision/plankton_transfer .

Automatic video analysis and classification of sleep-related hypermotor seizures and disorders of arousal.

OBJECTIVE: Sleep-related hypermotor epilepsy (SHE) is a focal epilepsy with seizures occurring mostly during sleep. SHE seizures present different motor characteristics ranging from dystonic posturing to hyperkinetic motor patterns, sometimes associated with affective symptoms and complex behaviors. Disorders of arousal (DOA) are sleep disorders with paroxysmal episodes that may present analogies with SHE seizures. Accurate interpretation of the different SHE patterns and their differentiation from DOA manifestations can be difficult and expensive, and can require highly skilled personnel not always available. Furthermore, it is operator dependent. METHODS: Common techniques for human motion analysis, such as wearable sensors (e.g., accelerometers) and motion capture systems, have been considered to overcome these problems. Unfortunately, these systems are cumbersome and they require trained personnel for marker and sensor positioning, limiting their use in the epilepsy domain. To overcome these problems, recently significant effort has been spent in studying automatic methods based on video analysis for the characterization of human motion. Systems based on computer vision and deep learning have been exploited in many fields, but epilepsy has received limited attention. RESULTS: In this paper, we present a pipeline composed of a set of three-dimensional convolutional neural networks that, starting from video recordings, reached an overall accuracy of 80% in the classification of different SHE semiology patterns and DOA. SIGNIFICANCE: The preliminary results obtained in this study highlight that our deep learning pipeline could be used by physicians as a tool to support them in the differential diagnosis of the different patterns of SHE and DOA, and encourage further investigation.

A markerless pipeline to analyze spontaneous movements of preterm infants.

BACKGROUND AND OBJECTIVE: The analysis of spontaneous movements of preterm infants is important because anomalous motion patterns can be a sign of neurological disorders caused by lesions in the developing brain. A diagnosis in the first weeks of child's life is crucial to plan timely and appropriate rehabilitative interventions. An accurate visual assessment of infants' spontaneous movements requires highly specialized personnel, not always available, and it is operator dependent. Motion capture systems, markers and wearable sensors are commonly used for human motion analysis, but they can be cumbersome, limiting their use in the study of infants' movements. METHODS: In this paper we propose a computer-aided pipeline to characterize and classify infants' motion from 2D video recordings. The final goal is detecting anomalous motion patterns. The implemented pipeline is based on computer vision and machine learning algorithms and includes a specific step to increase the interpretability of the results. Specifically, it can be summarized by the following steps: (i) body keypoints detection: we rely on a deep learning-based semantic features detector to localize the positions of meaningful landmark points on infants' bodies; (ii) parameters extraction: starting from the trajectories of the detected landmark points, we extract quantitative parameters describing infants motion patterns; (iii) classification: we implement different classifiers (Support Vector Machines, Random Forest, fully connected Neural Network, Long Short Term Memory) that, starting from the motion parameters, classify between normal or abnormal motion patterns. RESULTS: We tested the proposed pipeline on a dataset, recorded at the 40th gestational week, of 142 infants, 59 with evidence of neuromotor disorders according to a medical assessment carried out a posteriori. Our procedure successfully discriminates normal and anomalous motion patterns with a maximum accuracy of 85.7%. CONCLUSIONS: In conclusion, our pipeline has the potential to be adopted as a tool to support the early detection of abnormal motion patterns in preterm infants.

A semi-automatic toolbox for markerless effective semantic feature extraction.

VisionTool is an open-source python toolbox for semantic features extraction, capable to provide accurate features detectors for different applications, including motion analysis, markerless pose estimation, face recognition and biological cell tracking. VisionTool leverages transfer-learning with a large variety of deep neural networks allowing high-accuracy features detection with few training data. The toolbox offers a friendly graphical user interface, efficiently guiding the user through the entire process of features extraction. To facilitate broad usage and scientific community contribution, the code and a user guide are available at https://github.com/Malga-Vision/VisionTool.git .

Convolutional autoencoder based model HistoCAE for segmentation of viable tumor regions in liver whole-slide images.

Liver cancer is one of the leading causes of cancer deaths in Asia and Africa. It is caused by the Hepatocellular carcinoma (HCC) in almost 90% of all cases. HCC is a malignant tumor and the most common histological type of the primary liver cancers. The detection and evaluation of viable tumor regions in HCC present an important clinical significance since it is a key step to assess response of chemoradiotherapy and tumor cell proportion in genetic tests. Recent advances in computer vision, digital pathology and microscopy imaging enable automatic histopathology image analysis for cancer diagnosis. In this paper, we present a multi-resolution deep learning model HistoCAE for viable tumor segmentation in whole-slide liver histopathology images. We propose convolutional autoencoder (CAE) based framework with a customized reconstruction loss function for image reconstruction, followed by a classification module to classify each image patch as tumor versus non-tumor. The resulting patch-based prediction results are spatially combined to generate the final segmentation result for each WSI. Additionally, the spatially organized encoded feature map derived from small image patches is used to compress the gigapixel whole-slide images. Our proposed model presents superior performance to other benchmark models with extensive experiments, suggesting its efficacy for viable tumor area segmentation with liver whole-slide images.

Identification of excitatory-inhibitory links and network topology in large-scale neuronal assemblies from multi-electrode recordings.

Functional-effective connectivity and network topology are nowadays key issues for studying brain physiological functions and pathologies. Inferring neuronal connectivity from electrophysiological recordings presents open challenges and unsolved problems. In this work, we present a cross-correlation based method for reliably estimating not only excitatory but also inhibitory links, by analyzing multi-unit spike activity from large-scale neuronal networks. The method is validated by means of realistic simulations of large-scale neuronal populations. New results related to functional connectivity estimation and network topology identification obtained by experimental electrophysiological recordings from high-density and large-scale (i.e., 4096 electrodes) microtransducer arrays coupled to in vitro neural populations are presented. Specifically, we show that: (i) functional inhibitory connections are accurately identified in in vitro cortical networks, providing that a reasonable firing rate and recording length are achieved; (ii) small-world topology, with scale-free and rich-club features are reliably obtained, on condition that a minimum number of active recording sites are available. The method and procedure can be directly extended and applied to in vivo multi-units brain activity recordings.

SPICODYN: A Toolbox for the Analysis of Neuronal Network Dynamics and Connectivity from Multi-Site Spike Signal Recordings.

We implemented an automated and efficient open-source software for the analysis of multi-site neuronal spike signals. The software package, named SPICODYN, has been developed as a standalone windows GUI application, using C# programming language with Microsoft Visual Studio based on .NET framework 4.5 development environment. Accepted input data formats are HDF5, level 5 MAT and text files, containing recorded or generated time series spike signals data. SPICODYN processes such electrophysiological signals focusing on: spiking and bursting dynamics and functional-effective connectivity analysis. In particular, for inferring network connectivity, a new implementation of the transfer entropy method is presented dealing with multiple time delays (temporal extension) and with multiple binary patterns (high order extension). SPICODYN is specifically tailored to process data coming from different Multi-Electrode Arrays setups, guarantying, in those specific cases, automated processing. The optimized implementation of the Delayed Transfer Entropy and the High-Order Transfer Entropy algorithms, allows performing accurate and rapid analysis on multiple spike trains from thousands of electrodes.

Sphingomyelin as a myelin biomarker in CSF of acquired demyelinating neuropathies.

Fast, accurate and reliable methods to quantify the amount of myelin still lack, both in humans and experimental models. The overall objective of the present study was to demonstrate that sphingomyelin (SM) in the cerebrospinal fluid (CSF) of patients affected by demyelinating neuropathies is a myelin biomarker. We found that SM levels mirror both peripheral myelination during development and small myelin rearrangements in experimental models. As in acquired demyelinating peripheral neuropathies myelin breakdown occurs, SM amount in the CSF of these patients might detect the myelin loss. Indeed, quantification of SM in 262 neurological patients showed a significant increase in patients with peripheral demyelination (p = 3.81 * 10 - 8) compared to subjects affected by non-demyelinating disorders. Interestingly, SM alone was able to distinguish demyelinating from axonal neuropathies and differs from the principal CSF indexes, confirming the novelty of this potential CSF index. In conclusion, SM is a specific and sensitive biomarker to monitor myelin pathology in the CSF of peripheral neuropathies. Most importantly, SM assay is simple, fast, inexpensive, and promising to be used in clinical practice and drug development.

ToolConnect: A Functional Connectivity Toolbox for In vitro Networks.

Nowadays, the use of in vitro reduced models of neuronal networks to investigate the interplay between structural-functional connectivity and the emerging collective dynamics is a widely accepted approach. In this respect, a relevant advance for this kind of studies has been given by the recent introduction of high-density large-scale Micro-Electrode Arrays (MEAs) which have favored the mapping of functional connections and the recordings of the neuronal electrical activity. Although, several toolboxes have been implemented to characterize network dynamics and derive functional links, no specifically dedicated software for the management of huge amount of data and direct estimation of functional connectivity maps has been developed. toolconnect offers the implementation of up to date algorithms and a user-friendly Graphical User Interface (GUI) to analyze recorded data from large scale networks. It has been specifically conceived as a computationally efficient open-source software tailored to infer functional connectivity by analyzing the spike trains acquired from in vitro networks coupled to MEAs. In the current version, toolconnect implements correlation- (cross-correlation, partial-correlation) and information theory (joint entropy, transfer entropy) based core algorithms, as well as useful and practical add-ons to visualize functional connectivity graphs and extract some topological features. In this work, we present the software, its main features and capabilities together with some demonstrative applications on hippocampal recordings.

From functional to structural connectivity using partial correlation in neuronal assemblies.

OBJECTIVE: Our goal is to re-introduce an optimized version of the partial correlation to infer structural connections from functional-effective ones in dissociated neuronal cultures coupled to microelectrode arrays. APPROACH: We first validate our partialization procedure on in silico networks, mimicking different experimental conditions (i.e., different connectivity degrees and number of nodes) and comparing the partial correlation's performance with two gold-standard methods: cross-correlation and transfer entropy. Afterwards, to infer the structural connections in in vitro neuronal networks where the ground truth is unknown, we propose a thresholding heuristic approach. Then, to validate whether the partialization process correctly reconstructs macroscopic features of the network structure, we extract a modularity index from segregated in silico and in vitro models. Finally, as a case study, we apply our partialization procedure to analyze connectivity and topology on spontaneous developing and electrically stimulated in vitro cultures. MAIN RESULTS: In simulated networks, partial correlation outperforms cross-correlation and transfer entropy at low and medium connectivity degrees, not only in relatively small (60 nodes) but also in larger (120-240 nodes) assemblies. Furthermore, partial correlation correctly identifies interconnected neuronal sub-populations and allows one to derive network topology in in vitro cortical networks. SIGNIFICANCE: Our results support the idea that partial correlation is a good method for connectivity studies and can be applied to derive topological and structural features of neuronal assemblies.

A new connectivity toolbox to infer topological features of in-vitro neural networks.

A detailed analysis of functional connectivity of in vitro neural networks, as well as the possibility to understand the interplay between topology, structure, function and dynamics, is very important for better understanding how the nervous system represents and stores the information. Thus, we developed an informatics toolbox to infer functional connectivity in in-vitro neuronal networks. To prove the validity of the software tool and to verify its performances, we used it to estimate topological metrics on mature hippocampal assemblies coupled to Micro-Electrode Arrays (MEAs).