Wild-type and mutant (G2019S) leucine-rich repeat kinase 2 (LRRK2) associate with subunits of the translocase of outer mitochondrial membrane (TOM) complex.

Leucine-rich repeat kinase 2 (LRRK2) is important in various cellular processes including mitochondrial homeostasis and mutations in this gene lead to Parkinson's disease (PD). However, the full spectrum of LRRK2's functions remain to be elucidated. The translocase of outer mitochondrial membrane (TOM) complex is essential for the import of almost all nuclear-encoded mitochondrial proteins and is fundamental for cellular survival. Using co-immunoprecipitation, super-resolution structured illumination microscopy (SR-SIM), and 3D virtual reality (VR) assisted co-localization analysis techniques we show that wild-type and mutant (G2019S) LRRK2 associate and co-localize with subunits of the TOM complex, either under basal (dimethyl sulfoxide, DMSO) or stress-induced (carbonyl cyanide m-chlorophenyl hydrazine, CCCP) conditions. Interestingly, LRRK2 interacted with TOM40 under both DMSO and CCCP conditions, and when the PD causing mutation, G2019S was introduced, the association was not altered. Moreover, overexpression of G2019S LRRK2 resulted in the formation of large, perinuclear aggregates that co-localized with the TOM complex. Taken together, this is the first study to show that both WT and mutant LRRK2 associate with the TOM complex subunits. These findings provide additional evidence for LRRK2's role in mitochondrial function which has important implications for its role in PD pathogenesis.

Virtual reality assisted microscopy data visualization and colocalization analysis.

BACKGROUND: Confocal microscopes deliver detailed three-dimensional data and are instrumental in biological analysis and research. Usually, this three-dimensional data is rendered as a projection onto a two-dimensional display. We describe a system for rendering such data using a modern virtual reality (VR) headset. Sample manipulation is possible by fully-immersive hand-tracking and also by means of a conventional gamepad. We apply this system to the specific task of colocalization analysis, an important analysis tool in biological microscopy. We evaluate our system by means of a set of user trials. RESULTS: The user trials show that, despite inaccuracies which still plague the hand tracking, this is the most productive and intuitive interface. The inaccuracies nevertheless lead to a perception among users that productivity is low, resulting in a subjective preference for the gamepad. Fully-immersive manipulation was shown to be particularly effective when defining a region of interest (ROI) for colocalization analysis. CONCLUSIONS: Virtual reality offers an attractive and powerful means of visualization for microscopy data. Fully immersive interfaces using hand tracking show the highest levels of intuitiveness and consequent productivity. However, current inaccuracies in hand tracking performance still lead to a disproportionately critical user perception.

COVID-19 detection in cough, breath and speech using deep transfer learning and bottleneck features.

We present an experimental investigation into the effectiveness of transfer learning and bottleneck feature extraction in detecting COVID-19 from audio recordings of cough, breath and speech. This type of screening is non-contact, does not require specialist medical expertise or laboratory facilities and can be deployed on inexpensive consumer hardware such as a smartphone. We use datasets that contain cough, sneeze, speech and other noises, but do not contain COVID-19 labels, to pre-train three deep neural networks: a CNN, an LSTM and a Resnet50. These pre-trained networks are subsequently either fine-tuned using smaller datasets of coughing with COVID-19 labels in the process of transfer learning, or are used as bottleneck feature extractors. Results show that a Resnet50 classifier trained by this transfer learning process delivers optimal or near-optimal performance across all datasets achieving areas under the receiver operating characteristic (ROC AUC) of 0.98, 0.94 and 0.92 respectively for all three sound classes: coughs, breaths and speech. This indicates that coughs carry the strongest COVID-19 signature, followed by breath and speech. Our results also show that applying transfer learning and extracting bottleneck features using the larger datasets without COVID-19 labels led not only to improved performance, but also to a marked reduction in the standard deviation of the classifier AUCs measured over the outer folds during nested cross-validation, indicating better generalisation. We conclude that deep transfer learning and bottleneck feature extraction can improve COVID-19 cough, breath and speech audio classification, yielding automatic COVID-19 detection with a better and more consistent overall performance.

Automatic Non-Invasive Cough Detection based on Accelerometer and Audio Signals.

We present an automatic non-invasive way of detecting cough events based on both accelerometer and audio signals. The acceleration signals are captured by a smartphone firmly attached to the patient's bed, using its integrated accelerometer. The audio signals are captured simultaneously by the same smartphone using an external microphone. We have compiled a manually-annotated dataset containing such simultaneously-captured acceleration and audio signals for approximately 6000 cough and 68000 non-cough events from 14 adult male patients. Logistic regression (LR), support vector machine (SVM) and multilayer perceptron (MLP) classifiers provide a baseline and are compared with three deep architectures, convolutional neural network (CNN), long short-term memory (LSTM) network, and residual-based architecture (Resnet50) using a leave-one-out cross-validation scheme. We find that it is possible to use either acceleration or audio signals to distinguish between coughing and other activities including sneezing, throat-clearing, and movement on the bed with high accuracy. However, in all cases, the deep neural networks outperform the shallow classifiers by a clear margin and the Resnet50 offers the best performance, achieving an area under the ROC curve (AUC) exceeding 0.98 and 0.99 for acceleration and audio signals respectively. While audio-based classification consistently offers better performance than acceleration-based classification, we observe that the difference is very small for the best systems. Since the acceleration signal requires less processing power, and since the need to record audio is sidestepped and thus privacy is inherently secured, and since the recording device is attached to the bed and not worn, an accelerometer-based highly accurate non-invasive cough detector may represent a more convenient and readily accepted method in long-term cough monitoring.

COVID-19 cough classification using machine learning and global smartphone recordings.

We present a machine learning based COVID-19 cough classifier which can discriminate COVID-19 positive coughs from both COVID-19 negative and healthy coughs recorded on a smartphone. This type of screening is non-contact, easy to apply, and can reduce the workload in testing centres as well as limit transmission by recommending early self-isolation to those who have a cough suggestive of COVID-19. The datasets used in this study include subjects from all six continents and contain both forced and natural coughs, indicating that the approach is widely applicable. The publicly available Coswara dataset contains 92 COVID-19 positive and 1079 healthy subjects, while the second smaller dataset was collected mostly in South Africa and contains 18 COVID-19 positive and 26 COVID-19 negative subjects who have undergone a SARS-CoV laboratory test. Both datasets indicate that COVID-19 positive coughs are 15%-20% shorter than non-COVID coughs. Dataset skew was addressed by applying the synthetic minority oversampling technique (SMOTE). A leave-p-out cross-validation scheme was used to train and evaluate seven machine learning classifiers: logistic regression (LR), k-nearest neighbour (KNN), support vector machine (SVM), multilayer perceptron (MLP), convolutional neural network (CNN), long short-term memory (LSTM) and a residual-based neural network architecture (Resnet50). Our results show that although all classifiers were able to identify COVID-19 coughs, the best performance was exhibited by the Resnet50 classifier, which was best able to discriminate between the COVID-19 positive and the healthy coughs with an area under the ROC curve (AUC) of 0.98. An LSTM classifier was best able to discriminate between the COVID-19 positive and COVID-19 negative coughs, with an AUC of 0.94 after selecting the best 13 features from a sequential forward selection (SFS). Since this type of cough audio classification is cost-effective and easy to deploy, it is potentially a useful and viable means of non-contact COVID-19 screening.

Automatic cough classification for tuberculosis screening in a real-world environment.

Objective.The automatic discrimination between the coughing sounds produced by patients with tuberculosis (TB) and those produced by patients with other lung ailments.Approach.We present experiments based on a dataset of 1358 forced cough recordings obtained in a developing-world clinic from 16 patients with confirmed active pulmonary TB and 35 patients suffering from respiratory conditions suggestive of TB but confirmed to be TB negative. Using nested cross-validation, we have trained and evaluated five machine learning classifiers: logistic regression (LR), support vector machines, k-nearest neighbour, multilayer perceptrons and convolutional neural networks.Main Results.Although classification is possible in all cases, the best performance is achieved using LR. In combination with feature selection by sequential forward selection, our best LR system achieves an area under the ROC curve (AUC) of 0.94 using 23 features selected from a set of 78 high-resolution mel-frequency cepstral coefficients. This system achieves a sensitivity of 93% at a specificity of 95% and thus exceeds the 90% sensitivity at 70% specificity specification considered by the World Health Organisation (WHO) as a minimal requirement for a community-based TB triage test.Significance.The automatic classification of cough audio sounds, when applied to symptomatic patients requiring investigation for TB, can meet the WHO triage specifications for the identification of patients who should undergo expensive molecular downstream testing. This makes it a promising and viable means of low cost, easily deployable frontline screening for TB, which can benefit especially developing countries with a heavy TB burden.

Mitochondrial event localiser (MEL) to quantitativelydescribe fission, fusion and depolarisation in the three-dimensional space.

Mitochondrial fission and fusion play an important role not only in maintaining mitochondrial homeostasis but also in preserving overall cellular viability. However, quantitative analysis based on the three-dimensional localisation of these highly dynamic mitochondrial events in the cellular context has not yet been accomplished. Moreover, it remains largely uncertain where in the mitochondrial network depolarisation is most likely to occur. We present the mitochondrial event localiser (MEL), a method that allows high-throughput, automated and deterministic localisation and quantification of mitochondrial fission, fusion and depolarisation events in large three-dimensional microscopy time-lapse sequences. In addition, MEL calculates the number of mitochondrial structures as well as their combined and average volume for each image frame in the time-lapse sequence. The mitochondrial event locations can subsequently be visualised by superposition over the fluorescence micrograph z-stack. We apply MEL to both control samples as well as to cells before and after treatment with hydrogen peroxide (H2O2). An average of 9.3/7.2/2.3 fusion/fission/depolarisation events per cell were observed respectively for every 10 sec in the control cells. With peroxide treatment, the rate initially shifted toward fusion with and average of 15/6/3 events per cell, before returning to a new equilibrium not far from that of the control cells, with an average of 6.2/6.4/3.4 events per cell. These MEL results indicate that both pre-treatment and control cells maintain a fission/fusion equilibrium, and that depolarisation is higher in the post-treatment cells. When individually validating mitochondrial events detected with MEL, for a representative cell for the control and treated samples, the true-positive events were 47%/49%/14% respectively for fusion/fission/depolarisation events. We conclude that MEL is a viable method of quantitative mitochondrial event analysis.

Regression adjusted colocalisation colour mapping (RACC): A novel biological visual analysis method for qualitative colocalisation analysis of 3D fluorescence micrographs.

The qualitative analysis of colocalisation in fluorescence microscopy is of critical importance to the understanding of biological processes and cellular function. However, the degree of accuracy achieved may differ substantially when executing different yet commonly utilized colocalisation analyses. We propose a novel biological visual analysis method that determines the correlation within the fluorescence intensities and subsequently uses this correlation to assign a colourmap value to each voxel in a three-dimensional sample while also highlighting volumes with greater combined fluorescence intensity. This addresses the ambiguity and variability which can be introduced into the visualisation of the spatial distribution of correlation between two fluorescence channels when the colocalisation between these channels is not considered. Most currently employed and generally accepted methods of visualising colocalisation using a colourmap can be negatively affected by this ambiguity, for example by incorrectly indicating non-colocalised voxels as positively correlated. In this paper we evaluate the proposed method by applying it to both synthetic data and biological fluorescence micrographs and demonstrate how it can enhance the visualisation in a robust way by visualising only truly colocalised regions using a colourmap to indicate the qualitative measure of the correlation between the fluorescence intensities. This approach may substantially support fluorescence microscopy applications in which precise colocalisation analysis is of particular relevance.

A Comparative Study of Features for Acoustic Cough Detection Using Deep Architectures.

Automatic cough detection is key to tracking the condition of patients suffering from tuberculosis. We evaluate various acoustic features for performing cough detection using deep architectures. As most previous studies have adopted features designed for speech recognition, we assess the suitability of these techniques as well as their respective extraction parameters. Short-time Fourier transform (STFT), mel-frequency cepstral coefficients (MFCC) and mel-scaled filter banks (MFB) were evaluated using deep neural networks, convolutional neural networks and long-short term models. We find experimentally that, by regarding each cough sound as a single input feature instead of multiple shorter features, better performance can be achieved. Longer analysis windows also provide enhancement in contrast to the classic 25 ms frame. Although MFCC performance is improved by sinusoidal liftering, STFT and MFB lead to better results. Using MFB and the optimum segment and frame lengths, an improvement exceeding 7% in the area under the receiver operating characteristic curve across all classifiers is achieved.

Computer simulation and physical phantom models for estimating the dielectric properties of rhinoceros tissue.

In vivo and ex vivo sensors have the potential to aid tracking and anti-poaching endeavours and provide new insights into rhinoceros physiology and environment. However, the propagation of electromagnetic signals in rhinoceros tissue is currently not known. We present simulation and agar models of the rhinoceros that allow the investigation of electromagnetic propagation by in vivo and ex vivo devices without the need for surgery. Since the dielectric properties of rhinoceros tissue have not been documented, the conductivity and permittivity of the skin, fat, muscle, blood and other organs are first approximated by means of a meta-analysis that includes animals with similar physical properties. Subsequently, we develop anatomical models that include dermal layers, internal organs and a skeleton. We also develop a flank model that serves as an approximation of the anatomical model in certain situations. These models are used to determine the viability of communication between an in vivo device and an ex vivo device attached to the hind leg of the animal. Two types of antenna (microstrip-fed planar elliptical monopole antenna and printed inverted-F antenna) and three feasible implant locations (back, neck and chest) are considered. In addition to the computer models, phantom recipes using salt, sugar and agar are developed to match the dielectric properties of each tissue type at the industrial, scientific and medical (ISM) frequencies of 403MHz, 910MHz and 2.4GHz. The average error between the measured and theoretically predicted dielectric values was 6.22% over all recipes and 4.49% for the 2.4 GHz recipe specifically. When considering the predicted efficiency of the transmitting and receiving antennas, an agreement of 67.38% was demonstrated between the computer simulations and laboratory measurements using the agar rhinoceros flank models. Computer simulations using the anatomical model of the rhinoceros indicate that the chest is the optimal implant location and that best signal propagation is achieved using the planar inverted-F antenna (PIFA). Using this configuration, the simulations indicate that communication between the implant and an ex vivo device attached to the hind leg is challenging but possible. Furthermore, we find that the inclusion of factors such as the density and temperature of the phantom materials were found to be critical to the achievement of good agreement between practice and simulation.

Improved region of interest selection and colocalization analysis in three-dimensional fluorescence microscopy samples using virtual reality.

Although modern fluorescence microscopy produces detailed three-dimensional (3D) datasets, colocalization analysis and region of interest (ROI) selection is most commonly performed two-dimensionally (2D) using maximum intensity projections (MIP). However, these 2D projections exclude much of the available data. Furthermore, 2D ROI selections cannot adequately select complex 3D structures which may inadvertently lead to either the exclusion of relevant or the inclusion of irrelevant data points, consequently affecting the accuracy of the colocalization analysis. Using a virtual reality (VR) enabled system, we demonstrate that 3D visualization, sample interrogation and analysis can be achieved in a highly controlled and precise manner. We calculate several key colocalization metrics using both 2D and 3D derived super-resolved structured illumination-based data sets. Using a neuronal injury model, we investigate the change in colocalization between Tau and acetylated α-tubulin at control conditions, after 6 hours and again after 24 hours. We demonstrate that performing colocalization analysis in 3D enhances its sensitivity, leading to a greater number of statistically significant differences than could be established when using 2D methods. Moreover, by carefully delimiting the 3D structures under analysis using the 3D VR system, we were able to reveal a time dependent loss in colocalization between the Tau and microtubule network as an early event in neuronal injury. This behavior could not be reliably detected using a 2D based projection. We conclude that, using 3D colocalization analysis, biologically relevant samples can be interrogated and assessed with greater precision, thereby better exploiting the potential of fluorescence-based image analysis in biomedical research.

Clustering Acoustic Segments Using Multi-Stage Agglomerative Hierarchical Clustering.

Agglomerative hierarchical clustering becomes infeasible when applied to large datasets due to its O(N2) storage requirements. We present a multi-stage agglomerative hierarchical clustering (MAHC) approach aimed at large datasets of speech segments. The algorithm is based on an iterative divide-and-conquer strategy. The data is first split into independent subsets, each of which is clustered separately. Thus reduces the storage required for sequential implementations, and allows concurrent computation on parallel computing hardware. The resultant clusters are merged and subsequently re-divided into subsets, which are passed to the following iteration. We show that MAHC can match and even surpass the performance of the exact implementation when applied to datasets of speech segments.