Deep learning has proven to be one of the most effective methods in analyzing biological images to extract parameters fundamental for studying physiological functions and pathological conditions. In particular, when coupled with time-lapse microscopy (TLM), deep learning proves particularly effective in studying behaviors involving temporal dynamics. However, TLM videos are often affected by experimental noise and setup limitations, which can lead to inaccurate and poorly reproducible results. Taking advantage of the variational and generative capabilities of Variational Autoencoders (VAEs), we propose VAE-MOTION, a deep learning-based model for the analysis of cardiac contractile dynamics. By incorporating a temporal encoder into its architecture, our model allows the restoration of video quality by removing noise or increasing resolution, while simultaneously extracting accurate contraction-related signals from the latent space. The generation of synthetic videos allowed extensive training of VAE-MOTION, which subsequently validated on real videos from two different cardiac tissue models: 2D monolayers and 3D microtissues. VAE-MOTION was compared to two gold-standard methods in extracting contraction parameters relevant to drug efficacy or toxicity studies, demonstrating its potential for analyzing temporal dynamics in a given phenomenon or process.

The Anti-Nuclear Antibodies (ANA) test using Human Epithelial type 2 (HEp-2) cells in the Indirect Immuno-Fluorescence (IIF) assay protocol is considered the gold standard for detecting Connective Tissue Diseases. Computer-assisted systems for HEp-2 image analysis represent a growing field that harnesses the potential offered by novel machine learning techniques to address the classification of HEp-2 images and ANA patterns. In this study, we introduce an innovative platform based on transfer learning with pre-trained deep learning models. This platform combines the power of unsupervised deep description of HEp-2 images, a novel feature selection approach designed for unbalanced datasets, and independent testing using two distinct datasets from different hospitals to tackle cross-hardware compatibility issues. To enhance the trustworthiness of our method, we also present a modified version of gradient-weighted class activation mapping for regional explainability and introduce a new sample quality index based on the Jensen-Shannon divergence to enhance method reliability and quantify sample heterogeneity. The results we provide demonstrate exceptionally high performance in intensity and ANA pattern recognition when compared to state-of-the-art approaches. Our method's ability to eliminate the need for cell segmentation in favor of statistical analysis of the sample makes it applicable, robust, and versatile. Our future work will focus on addressing the challenge of mitotic spindle recognition by expanding our proposed approach to cover mixed patterns. • A novel AI platform for the classification of HEp-2 specimens without cell segmentation • Loss-based Multiclass Area Under the Curve feature selection approach • A novel Grad-CAM explainability method for transfer learning using pre-trained DL architectures • Novel sample quality index based on the Jensen-Shannon divergence index

MOTIVATION: In fluorescence microscopy, single-molecule localization microscopy (SMLM) techniques aim at localizing with high-precision high-density fluorescent molecules by stochastically activating and imaging small subsets of blinking emitters. Super resolution plays an important role in this field since it allows to go beyond the intrinsic light diffraction limit. RESULTS: In this work, we propose a deep learning-based algorithm for precise molecule localization of high-density frames acquired by SMLM techniques whose ℓ2-based loss function is regularized by non-negative and ℓ0-based constraints. The ℓ0 is relaxed through its continuous exact ℓ0 (CEL0) counterpart. The arising approach, named DeepCEL0, is parameter-free, more flexible, faster and provides more precise molecule localization maps if compared to the other state-of-the-art methods. We validate our approach on both simulated and real fluorescence microscopy data. AVAILABILITY AND IMPLEMENTATION: DeepCEL0 code is freely accessible at https://github.com/sedaboni/DeepCEL0.

We propose novel memristive gas sensor architectures that can significantly reduce process and parametric variations in a predictable manner, while improving accuracy and overall power consumption. The proposed architecture can also be configured to realize multifunction logic operations as well as Complementary Resistive Switch with low hardware overhead in the absence of gasses. Our results show that the proposed architecture is significantly immune to process and parametric effects compared to a single sensor and almost unaffected by wire resistance, while offering much higher accuracy and much lower power consumption compared to existing techniques.

Abstract In this paper, we present the novel Deep-MEG approach in which image-based representations of magnetoencephalography (MEG) data are combined with ensemble classifiers based on deep convolutional neural networks. For the scope of predicting the early signs of Alzheimer’s disease (AD), functional connectivity (FC) measures between the brain bio-magnetic signals originated from spatially separated brain regions are used as MEG data representations for the analysis. After stacking the FC indicators relative to different frequency bands into multiple images, a deep transfer learning model is used to extract different sets of deep features and to derive improved classification ensembles. The proposed Deep-MEG architectures were tested on a set of resting-state MEG recordings and their corresponding magnetic resonance imaging scans, from a longitudinal study involving 87 subjects. Accuracy values of 89% and 87% were obtained, respectively, for the early prediction of AD conversion in a sample of 54 mild cognitive impairment subjects and in a sample of 87 subjects, including 33 healthy controls. These results indicate that the proposed Deep-MEG approach is a powerful tool for detecting early alterations in the spectral–temporal connectivity profiles and in their spatial relationships.

The emerging tumor-on-chip (ToC) approaches allow to address biomedical questions out of reach with classical cell culture techniques: in biomimetic 3D hydrogels they partially reconstitute ex vivo the complexity of the tumor microenvironment and the cellular dynamics involving multiple cell types (cancer cells, immune cells, fibroblasts, etc.). However, a clear bottleneck is the extraction and interpretation of the rich biological information contained, sometime hidden, in the cell co-culture videos. In this work, we develop and apply novel video analysis algorithms to automatically measure the cytotoxic effects on human cancer cells (lung and breast) induced either by doxorubicin chemotherapy drug or by autologous tumor-infiltrating cytotoxic T lymphocytes (CTL). A live fluorescent dye (red) is used to selectively pre-stain the cancer cells before co-cultures and a live fluorescent reporter for caspase activity (green) is used to monitor apoptotic cell death. The here described open-source computational method, named STAMP (spatiotemporal apoptosis mapper), extracts the temporal kinetics and the spatial maps of cancer death, by localizing and tracking cancer cells in the red channel, and by counting the red to green transition signals, over 2-3 days. The robustness and versatility of the method is demonstrated by its application to different cell models and co-culture combinations. Noteworthy, this approach reveals the strong contribution of primary cancer-associated fibroblasts (CAFs) to breast cancer chemo-resistance, proving to be a powerful strategy to investigate intercellular cross-talks and drug resistance mechanisms. Moreover, we defined a new parameter, the 'potential of death induction', which is computed in time and in space to quantify the impact of dying cells on neighbor cells. We found that, contrary to natural death, cancer death induced by chemotherapy or by CTL is transmissible, in that it promotes the death of nearby cancer cells, suggesting the release of diffusible factors which amplify the initial cytotoxic stimulus.

This paper proposes a method to calculate the yield of a memristor based sensor array considered as the probability that the chip provides acceptable sensing results when the array is affected by manufacturing defects. The modeling is based on a Markov Chain approach, in which each state represents an operating chip configuration and the state transitions take into account manufacturing defects. The proposed method is applicable to evaluate the yield with different fault models to achieve the comparative yield obtained by several redundancy allocations.

Memristors are an attractive option for use in future architectures due to their non-volatility, high density and low power operation. Gas sensing is one of the proposed application of memristive devices. In spite of these advantages, memristors are susceptible to defect densities due to the nondeterministic nature of nano-scale fabrication. In this paper, a novel spice memristor model incorporating fault models that emulates the gas sensing behaviour with/without faults is developed for simulation and integration with design automation tools. Our simulation results show that the proposed non-linear model detects the presence of the oxidising/reducing gas and analyses the defects/faults affecting the functionality of the sensor.

Resistive memory, also known as memristor, is an emerging potential successor to traditional CMOS charge based memories. Memristors have also recently been proposed as a promising candidate for several additional applications such as logic design, sensing, non-volatile storage, neuromorphic computing, Physically Unclonable Functions (PUFs), Content-addressable memory (CAM) and reconfigurable computing. In this paper, we explore three unique applications of memristor technology based implementations, specifically from the perspective of sensing, logic, in-memory computing and their solutions. We review solar cell health monitoring and diagnosis, describe the proposed solutions, and provide directions in memristive gas sensing and in-memory computing. For the gas sensor application, in order to determine the number of memristors to ensure a certain level of accuracy in sensitivity, a technique to optimize the sensor array based on an acceptable sensitivity variation and minimum sensitivity margin is presented. These “out-of-the-box” emerging ideas for applications of memristive devices in enhancing robustness and, at the same time, how the requirements of robust design are enabling unconventional use of the devices. To this end, the papers considers some examples of this mutual interaction.

Gas sensing is one of the proposed application field of memristive devices. We used a crossbar array of memristors as gas sensor using the HP labs fabricated TiO2based memristor model in an attempt to improve sensing accuracy. We introduced the possibility of reliable multiple gases detection using multiple rows of memristors as separate sensor in a crossbar array. Our experimental results show that an array of memristors can minimise measurement errors as well as provide a good redundancy measure during gas sensing. Measurements taken from the sensors are also not affected by alternate current paths problem often experienced in crossbar architecture.

Research on automatic emotion recognition from speech has recently focused on the prediction of time-continuous dimensions (e.g., arousal and valence) of spontaneous and realistic expressions of emotion, as found in real-life interactions. However, the automatic prediction of such emotions poses several challenges, such as the subjectivity found in the definition of a gold-standard from a pool of raters and the issue of data scarcity in training models. In this work, we introduce a novel emotion recognition system, based on ensembles of single-speaker-regression-models. The estimation of emotion is provided by combining a subset of the initial pool of single-speaker-regression-models selecting those that are most concordant among them. The proposed approach allows the addition or removal of speakers from the ensemble without the necessity to re-build the entire recognition system. The simplicity of this aggregation strategy, coupled with the flexibility assured by the modular architecture, and the promising results observed on the RECOLA database highlight the potential implications of the proposed method in a real-life scenario and in particular in web-based applications.

Artificial olfaction systems have been investigated since more than two decades on alleged property of similarities between the receptive field of natural receptors and artificial sensors. Due to the limited number of sensors embedded in these systems the complexity of electronic noses is generally too low to allow the application of olfaction processing models. Actually the literature there are several models attempting to describe some olfaction functionalities, and the availability of an artificial platform to test models could be of great benefit for these studies. Recently, the use of optical image sensors has been demonstrated as a simple method to obtain large sensor arrays. Furthermore, an elegant and simple method to cluster individual sensors in classes allows for the definition of epithelium and glomerular layers. This system enables the application of a complex olfaction model, and these properties are here illustrated applying a glomerular compartmentalization model to the data generated by the exposure of such an artificial system to pure and mixed gases.

Artificial Olfaction (AO) data analysts have gained long term experience on nervous system based machine learning metaphors such as Artificial Neural Networks. In this work we propose and evaluate the use of a novel tool based on an emerging, however, powerful metaphor: the Artificial Immune Systems (AIS). AIS models were developed in the '90s; ever since they have reached significant maturity, and were to show good performance in both explorative data analysis and classification tasks. After selecting different artificial olfaction databases, we compare the utility of classic Back-Propagation Neural Network (BPNN) models with Artificial Immune Recognition Systems (AIRS) algorithms for classification problems, discussing its architectural strengths and weaknesses. Although BPNN retained a slight performance advantage on the investigated datasets, we were able to show that the AIS metaphor can express interesting characteristics for artificial olfaction data analysis. As an example, in a preliminary setup, the AIRS classifier showed superior performance when the sensor signals are affected by drift.

Although quartz microbalances (QMBs) are widely used, controlling their temperature is a challenge. Here a slightly modified quartz microbalance, driven by an original /spl Sigma//spl Delta/ electronic interface, acts as temperature sensor, flow sensor, heater, and resonator; no interference between the oscillator circuit and the /spl Sigma//spl Delta/ interface has been observed. In comparison with traditional solutions, the proposed approach is more accurate, faster, cheaper, more flexible, and more power efficient. As an additional, unique advantage of our system, the flow velocity is automatically measured by the QMB sensor itself. The technological and circuit solutions reported here are general and may be useful in other applications where temperature and/or flow velocity must be controlled.