ASAIO J. 2024 Sep 5. doi: 10.1097/MAT.0000000000002299. Online ahead of print.

ABSTRACT

Accurate activity classification is essential for the advancement of closed-loop control for left ventricular assist devices (LVADs), as it provides necessary feedback to adapt device operation to the patient's current state. Therefore, this study aims at using deep neural networks (DNNs) to precisely classify activity for these patients. Recordings from 13 LVAD patients were analyzed, including heart rate, LVAD flow, and accelerometer data, classifying activities into six states: active, inactive, lying, sitting, standing, and walking. Both binary and multiclass classifiers have been trained to distinguish between active and inactive states and to discriminate the remaining categories. The models were refined by testing several architectures, including recurrent and convolutional layers, optimized via hyperparameter search. Results demonstrate that integrating LVAD flow, heart rate, and accelerometer data leads to the highest accuracy in both binary and multiclass classification. The optimal architectures featured two and three bidirectional long short-term memory layers for binary and multiclass classifications, respectively, achieving accuracies of 91% and 84%. In this study, the potential of DNNs has been proven for providing a robust method for activity classification that is vital for the effective closed-loop control of medical devices in cardiac care.

PMID:39231213 | DOI:10.1097/MAT.0000000000002299

PLoS One. 2024 Sep 4;19(9):e0306385. doi: 10.1371/journal.pone.0306385. eCollection 2024.

ABSTRACT

Scanning Electron Microscope (SEM) is a crucial tool for studying microstructures of ceramic materials. However, the current practice heavily relies on manual efforts to extract porosity from SEM images. To address this issue, we propose PSTNet (Pyramid Segmentation Transformer Net) for grain and pore segmentation in SEM images, which merges multi-scale feature maps through operations like recombination and upsampling to predict and generate segmentation maps. These maps are used to predict the corresponding porosity at ceramic grain boundaries. To increase segmentation accuracy and minimize loss, we employ several strategies. (1) We train the micro-pore detection and segmentation model using publicly available Al2O3 and custom Y2O3 ceramic SEM images. We calculate the pixel percentage of segmented pores in SEM images to determine the surface porosity at the corresponding locations. (2) Utilizing high-temperature hot pressing sintering, we prepared and captured scanning electron microscope images of Y2O3 ceramics, with which a Y2O3 ceramic dataset was constructed through preprocessing and annotation. (3) We employed segmentation penalty cross-entropy loss, smooth L1 loss, and structural similarity (SSIM) loss as the constituent terms of a joint loss function. The segmentation penalty cross-entropy loss helps suppress segmentation loss bias, smooth L1 loss is utilized to reduce noise in images, and incorporating structural similarity into the loss function computation guides the model to better learn structural features of images, significantly improving the accuracy and robustness of semantic segmentation. (4) In the decoder stage, we utilized an improved version of the multi-head attention mechanism (MHA) for feature fusion, leading to a significant enhancement in model performance. Our model training is based on publicly available laser-sintered Al2O3 ceramic datasets and self-made high-temperature hot-pressed sintered Y2O3 ceramic datasets, and validation has been completed. Our Pix Acc score improves over the baseline by 12.2%, 86.52 vs. 76.01, and the mIoU score improves from by 25.5%, 69.10 vs. 51.49. The average relative errors on datasets Y2O3 and Al2O3 were 6.9% and 6.36%, respectively.

PMID:39231159 | DOI:10.1371/journal.pone.0306385

IEEE Trans Cybern. 2024 Sep 4;PP. doi: 10.1109/TCYB.2024.3447108. Online ahead of print.

ABSTRACT

Deep learning-based soft sensor modeling methods have been extensively studied and applied to industrial processes in the last decade. However, existing soft sensor models mainly focus on the current step prediction in real time and ignore the multistep prediction in advance. In actual industrial applications, compared to the current step prediction, it is more useful for on-site workers to predict some key performance indicators in advance. Nowadays, multistep prediction task still suffers from two key issues: 1) complex coupling relationships between process variables and 2) long-term dependency learning. To ravel out these two problems, in this article, we propose a graph-based time-frequency two-stream network to achieve multistep prediction. Specifically, a multigraph attention layer is proposed to model the dynamical coupling relationships between process variables from the graph perspective. Then, in the time-frequency two-stream network, multi-GAT is used to extract time-domain features and frequency-domain features for long-term dependency, respectively. Furthermore, we propose a feature fusion module to combine these two kinds of features based on the minimum redundancy and maximum correlation learning paradigm. Finally, extensive experiments on two real-world industrial datasets show that the proposed multistep prediction model outperforms the state-of-the-art models. In particular, compared to the existing SOTA method, the proposed method has achieved 12.40%, 22.49%, and 21.98% improvement in RMSE, MAE, and MAPE on the three-step prediction task using waste incineration dataset.

PMID:39231064 | DOI:10.1109/TCYB.2024.3447108

IEEE Trans Biomed Eng. 2024 Sep 4;PP. doi: 10.1109/TBME.2024.3454545. Online ahead of print.

ABSTRACT

OBJECTIVE: In Long-Term Monitoring (LTM), noise significantly impacts the quality of the electrocardiogram (ECG), posing challenges for accurate diagnosis and time-consuming analysis. The clinical severity of noise refers to the difficulty in interpreting the clinical content of the ECG, in contrast to the traditional approach based on quantitative severity. In a previous study, we trained Machine Learning (ML) algorithms using a data repository labeled according to the clinical severity. In this work, we explore Deep Learning (DL) models in the same database to design architectures that provide explainability of the decision making process.

METHODS: We have developed two sets of Convolutional Neural Networks (CNNs): a 1-D CNN model designed from scratch, and pre-trained 2-D CNNs fine-tuned through transfer learning. Additionally, we have designed two Autoencoder (AE) architectures to provide model interpretability by exploiting the data regionalization in the latent spaces.

RESULTS: The DL systems yield superior classification performance than the previous ML approaches, achieving an F1-score up to 0.84 in the test set considering patient separation to avoid intra-patient overfitting. The interpretable architectures have shown similar performance with the advantage of qualitative explanations.

CONCLUSIONS: The integration of DL and interpretable systems has proven to be highly effective in classifying clinical noise in LTM ECG recordings. This approach can enhance clinicians' confidence in clinical decision support systems based on learning methods, a key point for this technology transfer.

SIGNIFICANCE: The proposed systems can help healthcare professionals to discriminate the parts of the ECG that contain valuable information to provide a diagnosis.

PMID:39231059 | DOI:10.1109/TBME.2024.3454545

Interdiscip Sci. 2024 Sep 4. doi: 10.1007/s12539-024-00638-7. Online ahead of print.

ABSTRACT

Using genes which have been experimentally-validated for diseases (functions) can develop machine learning methods to predict new disease/function-genes. However, the prediction of both function-genes and disease-genes faces the same problem: there are only certain positive examples, but no negative examples. To solve this problem, we proposed a function/disease-genes prediction algorithm based on network embedding (Variational Graph Auto-Encoders, VGAE) and one-class classification (Fast Minimum Covariance Determinant, Fast-MCD): VGAEMCD. Firstly, we constructed a protein-protein interaction (PPI) network centered on experimentally-validated genes; then VGAE was used to get the embeddings of nodes (genes) in the network; finally, the embeddings were input into the improved deep learning one-class classifier based on Fast-MCD to predict function/disease-genes. VGAEMCD can predict function-gene and disease-gene in a unified way, and only the experimentally-verified genes are needed to provide (no need for expression profile). VGAEMCD outperforms classical one-class classification algorithms in Recall, Precision, F-measure, Specificity, and Accuracy. Further experiments show that seven metrics of VGAEMCD are higher than those of state-of-art function/disease-genes prediction algorithms. The above results indicate that VGAEMCD can well learn the distribution characteristics of positive examples and accurately identify function/disease-genes.

PMID:39230798 | DOI:10.1007/s12539-024-00638-7

Interdiscip Sci. 2024 Sep 4. doi: 10.1007/s12539-024-00641-y. Online ahead of print.

ABSTRACT

BACKGROUND: Accurate identification of cancer subtypes is crucial for disease prognosis evaluation and personalized patient management. Recent advances in computational methods have demonstrated that multi-omics data provides valuable insights into tumor molecular subtyping. However, the high dimensionality and small sample size of the data may result in ambiguous and overlapping cancer subtypes during clustering. In this study, we propose a novel contrastive-learning-based approach to address this issue. The proposed end-to-end deep learning method can extract crucial information from the multi-omics features by self-supervised learning for patient clustering.

RESULTS: By applying our method to nine public cancer datasets, we have demonstrated superior performance compared to existing methods in separating patients with different survival outcomes (p < 0.05). To further evaluate the impact of various omics data on cancer survival, we developed an XGBoost classification model and found that mRNA had the highest importance score, followed by DNA methylation and miRNA. In the presented case study, our method successfully clustered subtypes and identified 14 cancer-related genes, of which 12 (85.7%) were validated through literature review.

CONCLUSIONS: Our findings demonstrate that our method is capable of identifying cancer subtypes that are both statistically and biologically significant. The code about COLCS is given at: https://github.com/Mercuriiio/COLCS .

PMID:39230797 | DOI:10.1007/s12539-024-00641-y

Neurosurg Rev. 2024 Sep 4;47(1):535. doi: 10.1007/s10143-024-02803-4.

ABSTRACT

Spine surgery is essential for restoring alignment, stability, and function in patients with cervical spine injuries, especially when instability, pain, deformity, or progressive nerve damage is present. Effective wound closure is vital in these procedures, aiming to promote rapid healing, reduce infection risks, enable early mobilization, and ensure satisfactory cosmetic results. However, there is limited evidence on the optimal wound closure technique for posterior spine surgery, highlighting the need for innovative approaches. A study by Glener et al. evaluated the effectiveness of STRATAFIX™ Symmetric barbed sutures compared to traditional braided absorbable sutures in spinal surgery. In a randomized trial involving 20 patients, the STRATAFIX™ group demonstrated a shorter mean closure time and significantly fewer sutures used, though without a statistically significant reduction in closure time. No significant differences were observed in postoperative complications between the groups during a six-month follow-up. While the findings suggest potential cost savings and efficiency improvements with STRATAFIX™, the study's small sample size and short follow-up period limit its generalizability. Furthermore, AI-based models, such as the Xception deep learning model, show promise in improving suture training accuracy for medical students, which could enhance surgical outcomes and reduce complications. Despite the promising results, further research with larger sample sizes, extended follow-up periods, and multi-center trials is necessary to validate the effectiveness of barbed sutures like STRATAFIX™ in neurosurgery. The integration of AI in surgical training and continued exploration of innovative techniques are essential to advancing the field and optimizing patient care in spinal surgery.

PMID:39230765 | DOI:10.1007/s10143-024-02803-4

Bioinformatics. 2024 Sep 1;40(Supplement_2):ii45-ii52. doi: 10.1093/bioinformatics/btae402.

ABSTRACT

MOTIVATION: Dictionary-based named entity recognition (NER) allows terms to be detected in a corpus and normalized to biomedical databases and ontologies. However, adaptation to different entity types requires new high-quality dictionaries and associated lists of blocked names for each type. The latter are so far created by identifying cases that cause many false positives through manual inspection of individual names, a process that scales poorly.

RESULTS: In this work, we aim to improve block list s by automatically identifying names to block, based on the context in which they appear. By comparing results of three well-established biomedical NER methods, we generated a dataset of over 12.5 million text spans where the methods agree on the boundaries and type of entity tagged. These were used to generate positive and negative examples of contexts for four entity types (genes, diseases, species, and chemicals), which were used to train a Transformer-based model (BioBERT) to perform entity type classification. Application of the best model (F1-score = 96.7%) allowed us to generate a list of problematic names that should be blocked. Introducing this into our system doubled the size of the previous list of corpus-wide blocked names. In addition, we generated a document-specific list that allows ambiguous names to be blocked in specific documents. These changes boosted text mining precision by ∼5.5% on average, and over 8.5% for chemical and 7.5% for gene names, positively affecting several biological databases utilizing this NER system, like the STRING database, with only a minor drop in recall (0.6%).

AVAILABILITY AND IMPLEMENTATION: All resources are available through Zenodo https://doi.org/10.5281/zenodo.11243139 and GitHub https://doi.org/10.5281/zenodo.10289360.

PMID:39230709 | DOI:10.1093/bioinformatics/btae402

Bioinformatics. 2024 Sep 1;40(Supplement_2):ii4-ii10. doi: 10.1093/bioinformatics/btae403.

ABSTRACT

With the development of high-throughput technologies, genomics datasets rapidly grow in size, including functional genomics data. This has allowed the training of large Deep Learning (DL) models to predict epigenetic readouts, such as protein binding or histone modifications, from genome sequences. However, large dataset sizes come at a price of data consistency, often aggregating results from a large number of studies, conducted under varying experimental conditions. While data from large-scale consortia are useful as they allow studying the effects of different biological conditions, they can also contain unwanted biases from confounding experimental factors. Here, we introduce Metadata-guided Feature Disentanglement (MFD)-an approach that allows disentangling biologically relevant features from potential technical biases. MFD incorporates target metadata into model training, by conditioning weights of the model output layer on different experimental factors. It then separates the factors into disjoint groups and enforces independence of the corresponding feature subspaces with an adversarially learned penalty. We show that the metadata-driven disentanglement approach allows for better model introspection, by connecting latent features to experimental factors, without compromising, or even improving performance in downstream tasks, such as enhancer prediction, or genetic variant discovery. The code will be made available at https://github.com/HealthML/MFD.

PMID:39230700 | DOI:10.1093/bioinformatics/btae403

Bioinformatics. 2024 Sep 1;40(Supplement_2):ii128-ii136. doi: 10.1093/bioinformatics/btae395.

ABSTRACT

SUMMARY: Single-neuron morphology, the study of the structure, form, and shape of a group of specialized cells in the nervous system, is of vital importance to define the type of neurons, assess changes in neuronal development and aging and determine the effects of brain disorders and treatments. Despite the recent surge in the amount of available neuron morphology reconstructions due to advancements in microscopy imaging, existing computational and deep learning methods for modeling neuron morphology have been limited in both scale and accuracy. In this paper, we propose MorphRep, a model for learning meaningful representation of neuron morphology pre-trained with over 250 000 existing neuron morphology data. By encoding the neuron morphology into graph-structured data, using graph transformers for feature encoding and enforcing the consistency between multiple augmented views of neuron morphology, MorphRep achieves the state of the art performance on widely used benchmarking datasets. Meanwhile, MorphRep can accurately characterize the neuron morphology space across neuron morphometrics, fine-grained cell types, brain regions and ages. Furthermore, MorphRep can be applied to distinguish neurons under a wide range of conditions, including genetic perturbation, drug injection, environment change and disease. In summary, MorphRep provides an effective strategy to embed and represent neuron morphology and can be a valuable tool in integrating cell morphology into single-cell multiomics analysis.

AVAILABILITY AND IMPLEMENTATION: The codebase has been deposited in https://github.com/YaxuanLi-cn/MorphRep.

PMID:39230697 | DOI:10.1093/bioinformatics/btae395

Bioinformatics. 2024 Sep 1;40(Supplement_2):ii182-ii189. doi: 10.1093/bioinformatics/btae388.

ABSTRACT

MOTIVATION: Cancer is a very heterogeneous disease that can be difficult to treat without addressing the specific mechanisms driving tumour progression in a given patient. High-throughput screening and sequencing data from cancer cell-lines has driven many developments in drug development, however, there are important aspects crucial to precision medicine that are often overlooked, namely the inherent differences between tumours in patients and the cell-lines used to model them in vitro. Recent developments in transfer learning methods for patient and cell-line data have shown progress in translating results from cell-lines to individual patients in silico. However, transfer learning can be forceful and there is a risk that clinically relevant patterns in the omics profiles of patients are lost in the process.

RESULTS: We present MODAE, a novel deep learning algorithm to integrate omics profiles from cell-lines and patients for the purposes of exploring precision medicine opportunities. MODAE implements patient survival prediction as an additional task in a drug-sensitivity transfer learning schema and aims to balance autoencoding, domain adaptation, drug-sensitivity prediction, and survival prediction objectives in order to better preserve the heterogeneity between patients that is relevant to survival. While burdened with these additional tasks, MODAE performed on par with baseline survival models, but struggled in the drug-sensitivity prediction task. Nevertheless, these preliminary results were promising and show that MODAE provides a novel AI-based method for prioritizing drug treatments for high-risk patients.

AVAILABILITY AND IMPLEMENTATION: https://github.com/UEFBiomedicalInformaticsLab/MODAE.

PMID:39230696 | DOI:10.1093/bioinformatics/btae388

Bioinformatics. 2024 Sep 1;40(Supplement_2):ii146-ii154. doi: 10.1093/bioinformatics/btae393.

ABSTRACT

SUMMARY: Measurement of single-cell gene expression at different timepoints enables the study of cell development. However, due to the resource constraints and technical challenges associated with the single-cell experiments, researchers can only profile gene expression at discrete and sparsely sampled timepoints. This missing timepoint information impedes downstream cell developmental analyses. We propose scNODE, an end-to-end deep learning model that can predict in silico single-cell gene expression at unobserved timepoints. scNODE integrates a variational autoencoder with neural ordinary differential equations to predict gene expression using a continuous and nonlinear latent space. Importantly, we incorporate a dynamic regularization term to learn a latent space that is robust against distribution shifts when predicting single-cell gene expression at unobserved timepoints. Our evaluations on three real-world scRNA-seq datasets show that scNODE achieves higher predictive performance than state-of-the-art methods. We further demonstrate that scNODE's predictions help cell trajectory inference under the missing timepoint paradigm and the learned latent space is useful for in silico perturbation analysis of relevant genes along a developmental cell path.

AVAILABILITY AND IMPLEMENTATION: The data and code are publicly available at https://github.com/rsinghlab/scNODE.

PMID:39230694 | DOI:10.1093/bioinformatics/btae393

Bioinformatics. 2024 Sep 1;40(Supplement_2):ii79-ii86. doi: 10.1093/bioinformatics/btae381.

ABSTRACT

MOTIVATION: For the alignment of large numbers of protein sequences, tools are predominant that decide to align two residues using only simple prior knowledge, e.g. amino acid substitution matrices, and using only part of the available data. The accuracy of state-of-the-art programs declines with decreasing sequence identity and when increasingly large numbers of sequences are aligned. Recently, transformer-based deep-learning models started to harness the vast amount of protein sequence data, resulting in powerful pretrained language models with the main purpose of generating high-dimensional numerical representations, embeddings, for individual sites that agglomerate evolutionary, structural, and biophysical information.

RESULTS: We extend the traditional profile hidden Markov model so that it takes as inputs unaligned protein sequences and the corresponding embeddings. We fit the model with gradient descent using our existing differentiable hidden Markov layer. All sequences and their embeddings are jointly aligned to a model of the protein family. We report that our upgraded HMM-based aligner, learnMSA2, combined with the ProtT5-XL protein language model aligns on average almost 6% points more columns correctly than the best amino acid-based competitor and scales well with sequence number. The relative advantage of learnMSA2 over other programs tends to be greater when the sequence identity is lower and when the number of sequences is larger. Our results strengthen the evidence on the rich information contained in protein language models' embeddings and their potential downstream impact on the field of bioinformatics. Availability and implementation: https://github.com/Gaius-Augustus/learnMSA, PyPI and Bioconda, evaluation: https://github.com/felbecker/snakeMSA.

PMID:39230690 | DOI:10.1093/bioinformatics/btae381

Eur Arch Otorhinolaryngol. 2024 Sep 4. doi: 10.1007/s00405-024-08948-8. Online ahead of print.

ABSTRACT

PURPOSE: To assess the accuracy of deep learning models for the diagnosis of maxillary fungal ball rhinosinusitis (MFB) and to compare the accuracy, sensitivity, specificity, precision, and F1-score with a rhinologist.

METHODS: Data from 1539 adult chronic rhinosinusitis (CRS) patients who underwent paranasal sinus computed tomography (CT) were collected. The overall dataset consisted of 254 MFB cases and 1285 non-MFB cases. The CT images were constructed and labeled to form the deep learning models. Seventy percent of the images were used for training the deep-learning models, and 30% were used for testing. Whole image analysis and instance segmentation analysis were performed using three different architectures: MobileNetv3, ResNet50, and ResNet101 for whole image analysis, and YOLOv5X-SEG, YOLOv8X-SEG, and YOLOv9-C-SEG for instance segmentation analysis. The ROC curve was assessed. Accuracy, sensitivity (recall), specificity, precision, and F1-score were compared between the models and a rhinologist. Kappa agreement was evaluated.

RESULTS: Whole image analysis showed lower precision, recall, and F1-score compared to instance segmentation. The models exhibited an area under the ROC curve of 0.86 for whole image analysis and 0.88 for instance segmentation. In the testing dataset for whole images, the MobileNet V3 model showed 81.00% accuracy, 47.40% sensitivity, 87.90% specificity, 66.80% precision, and a 67.20% F1 score. Instance segmentation yielded the best evaluation with YOLOv8X-SEG showing 94.10% accuracy, 85.90% sensitivity, 95.80% specificity, 88.90% precision, and an 89.80% F1-score. The rhinologist achieved 93.5% accuracy, 84.6% sensitivity, 95.3% specificity, 78.6% precision, and an 81.5% F1-score.

CONCLUSION: Utilizing paranasal sinus CT imaging with enhanced localization and constructive instance segmentation in deep learning models can be the practical promising deep learning system in assisting physicians for diagnosing maxillary fungal ball.

PMID:39230611 | DOI:10.1007/s00405-024-08948-8

Heliyon. 2024 Aug 10;10(16):e35959. doi: 10.1016/j.heliyon.2024.e35959. eCollection 2024 Aug 30.

ABSTRACT

The Pegon script is an Arabic-based writing system used for Javanese, Sundanese, Madurese, and Indonesian languages. Due to various reasons, this script is now mainly found among collectors and private Islamic boarding schools (pesantren), creating a need for its preservation. One preservation method is digitization through transcription into machine-encoded text, known as OCR (Optical Character Recognition). No published literature exists on OCR systems for this specific script. This research explores the OCR of Pegon typed manuscripts, introducing novel synthesized and real annotated datasets for this task. These datasets evaluate proposed OCR methods, especially those adapted from existing Arabic OCR systems. Results show that deep learning techniques outperform conventional ones, which fail to detect Pegon text. The proposed system uses YOLOv5 for line segmentation and a CTC-CRNN architecture for line text recognition, achieving an F1-score of 0.94 for segmentation and a CER of 0.03 for recognition.

PMID:39229500 | PMC:PMC11369439 | DOI:10.1016/j.heliyon.2024.e35959

bioRxiv [Preprint]. 2024 Aug 20:2024.08.19.608555. doi: 10.1101/2024.08.19.608555.

ABSTRACT

Abnormal DNA ploidy, found in numerous cancers, is increasingly being recognized as a contributor in driving chromosomal instability, genome evolution, and the heterogeneity that fuels cancer cell progression. Furthermore, it has been linked with poor prognosis of cancer patients. While next-generation sequencing can be used to approximate tumor ploidy, it has a high error rate for near-euploid states, a high cost and is time consuming, motivating alternative rapid quantification methods. We introduce PloiViT, a transformer-based model for tumor ploidy quantification that outperforms traditional machine learning models, enabling rapid and cost-effective quantification directly from pathology slides. We trained PloiViT on a dataset of fifteen cancer types from The Cancer Genome Atlas and validated its performance in multiple independent cohorts. Additionally, we explored the impact of self-supervised feature extraction on performance. PloiViT, using self-supervised features, achieved the lowest prediction error in multiple independent cohorts, exhibiting better generalization capabilities. Our findings demonstrate that PloiViT predicts higher ploidy values in aggressive cancer groups and patients with specific mutations, validating PloiViT potential as complementary for ploidy assessment to next-generation sequencing data. To further promote its use, we release our models as a user-friendly inference application and a Python package for easy adoption and use.

PMID:39229200 | PMC:PMC11370345 | DOI:10.1101/2024.08.19.608555

bioRxiv [Preprint]. 2024 Aug 19:2024.08.16.608331. doi: 10.1101/2024.08.16.608331.

ABSTRACT

Somatic variant detection is an integral part of cancer genomics analysis. While most methods have focused on short-read sequencing, long-read technologies now offer potential advantages in terms of repeat mapping and variant phasing. We present DeepSomatic, a deep learning method for detecting somatic SNVs and insertions and deletions (indels) from both short-read and long-read data, with modes for whole-genome and exome sequencing, and able to run on tumor-normal, tumor-only, and with FFPE-prepared samples. To help address the dearth of publicly available training and benchmarking data for somatic variant detection, we generated and make openly available a dataset of five matched tumor-normal cell line pairs sequenced with Illumina, PacBio HiFi, and Oxford Nanopore Technologies, along with benchmark variant sets. Across samples and technologies (short-read and long-read), DeepSomatic consistently outperforms existing callers, particularly for indels.

PMID:39229187 | PMC:PMC11370364 | DOI:10.1101/2024.08.16.608331

bioRxiv [Preprint]. 2024 Aug 24:2024.05.10.593561. doi: 10.1101/2024.05.10.593561.

ABSTRACT

Deep learning tools for behavior analysis have enabled important new insights and discoveries in neuroscience. Yet, they often compromise interpretability and generalizability for performance, making it difficult to quantitively compare phenotypes across datasets and research groups. We developed a novel deep learning-based behavior analysis pipeline, Avian Vocalization Network (AVN), for the learned vocalizations of the most extensively studied vocal learning model species - the zebra finch. AVN annotates songs with high accuracy across multiple animal colonies without the need for any additional training data and generates a comprehensive set of interpretable features to describe the syntax, timing, and acoustic properties of song. We use this feature set to compare song phenotypes across multiple research groups and experiments, and to predict a bird's stage in song development. Additionally, we have developed a novel method to measure song imitation that requires no additional training data for new comparisons or recording environments, and outperforms existing similarity scoring methods in its sensitivity and agreement with expert human judgements of song similarity. These tools are available through the open-source AVN python package and graphical application, which makes them accessible to researchers without any prior coding experience. Altogether, this behavior analysis toolkit stands to facilitate and accelerate the study of vocal behavior by enabling a standardized mapping of phenotypes and learning outcomes, thus helping scientists better link behavior to the underlying neural processes.

PMID:39229184 | PMC:PMC11370480 | DOI:10.1101/2024.05.10.593561

bioRxiv [Preprint]. 2024 Aug 19:2024.08.16.608247. doi: 10.1101/2024.08.16.608247.

ABSTRACT

The multiplexed immunofluorescence (mIF) platform enables biomarker discovery through the simultaneous detection of multiple markers on a single tissue slide, offering detailed insights into intratumor heterogeneity and the tumor-immune microenvironment at spatially resolved single cell resolution. However, current mIF image analyses are labor-intensive, requiring specialized pathology expertise which limits their scalability and clinical application. To address this challenge, we developed CellGate, a deep-learning (DL) computational pipeline that provides streamlined, end-to-end whole-slide mIF image analysis including nuclei detection, cell segmentation, cell classification, and combined immuno-phenotyping across stacked images. The model was trained on over 750,000 single cell images from 34 melanomas in a retrospective cohort of patients using whole tissue sections stained for CD3, CD8, CD68, CK-SOX10, PD-1, PD-L1, and FOXP3 with manual gating and extensive pathology review. When tested on new whole mIF slides, the model demonstrated high precision-recall AUC. Further validation on whole-slide mIF images of 9 primary melanomas from an independent cohort confirmed that CellGate can reproduce expert pathology analysis with high accuracy. We show that spatial immuno-phenotyping results using CellGate provide deep insights into the immune cell topography and differences in T cell functional states and interactions with tumor cells in patients with distinct histopathology and clinical characteristics. This pipeline offers a fully automated and parallelizable computing process with substantially improved consistency for cell type classification across images, potentially enabling high throughput whole-slide mIF tissue image analysis for large-scale clinical and research applications.

PMID:39229153 | PMC:PMC11370407 | DOI:10.1101/2024.08.16.608247

bioRxiv [Preprint]. 2024 Aug 21:2024.08.20.608885. doi: 10.1101/2024.08.20.608885.

ABSTRACT

Chromatin-sensitive Partial Wave Spectroscopic (csPWS) microscopy offers a non-invasive glimpse into the mass density distribution of cellular structures at the nanoscale, leveraging the spectroscopic information. Such capability allows us to analyze the chromatin structure and organization and the global transcriptional state of the cell nuclei for the study of its role in carcinogenesis. Accurate segmentation of the nuclei in csPWS microscopy images is an essential step in isolating them for further analysis. However, manual segmentation is error-prone, biased, time-consuming, and laborious, resulting in disrupted nuclear boundaries with partial or over-segmentation. Here, we present an innovative deep-learning-driven approach to automate the accurate nuclei segmentation of label-free live cell csPWS microscopy imaging data. Our approach, csPWS-seg, harnesses the Convolutional Neural Networks-based U-Net model with an attention mechanism to automate the accurate cell nuclei segmentation of csPWS microscopy images. We leveraged the structural, physical, and biological differences between the cytoplasm, nucleus, and nuclear periphery to construct three distinct csPWS feature images for nucleus segmentation. Using these images of HCT116 cells, csPWS-seg achieved superior performance with a median Intersection over Union (IoU) of 0.80 and a Dice Similarity Coefficient (DSC) score of 0.88. The csPWS-seg overcame the segmentation performance over the baseline U-Net model and another attention-based model, SE-U-Net, marking a significant improvement in segmentation accuracy. Further, we analyzed the performance of our proposed model with four loss functions: binary cross-entropy loss, focal loss, dice loss, and Jaccard loss. The csPWS-seg with focal loss provided the best results compared to other loss functions. The automatic and accurate nuclei segmentation offered by the csPWS-seg not only automates, accelerates, and streamlines csPWS data analysis but also enhances the reliability of subsequent chromatin analysis research, paving the way for more accurate diagnostics, treatment, and understanding of cellular mechanisms for carcinogenesis.

PMID:39229026 | PMC:PMC11370422 | DOI:10.1101/2024.08.20.608885