IEEE Trans Pattern Anal Mach Intell. 2024 Apr 12;PP. doi: 10.1109/TPAMI.2024.3388150. Online ahead of print.

ABSTRACT

Photometric stereo recovers the surface normals of an object from multiple images with varying shading cues, i.e., modeling the relationship between surface orientation and intensity at each pixel. Photometric stereo prevails in superior per-pixel resolution and fine reconstruction details. However, it is a complicated problem because of the non-linear relationship caused by non-Lambertian surface reflectance. Recently, various deep learning methods have shown a powerful ability in the context of photometric stereo against non-Lambertian surfaces. This paper provides a comprehensive review of existing deep learning-based calibrated photometric stereo methods utilizing orthographic cameras and directional light sources. We first analyze these methods from different perspectives, including input processing, supervision, and network architecture. We summarize the performance of deep learning photometric stereo models on the most widely-used benchmark data set. This demonstrates the advanced performance of deep learning-based photometric stereo methods. Finally, we give suggestions and propose future research trends based on the limitations of existing models.

PMID:38607717 | DOI:10.1109/TPAMI.2024.3388150

IEEE Trans Ultrason Ferroelectr Freq Control. 2024 Apr 12;PP. doi: 10.1109/TUFFC.2024.3388102. Online ahead of print.

ABSTRACT

Ultrasound localization microscopy (ULM) overcomes the acoustic diffraction limit by localizing tiny microbubbles (MBs), thus enabling the microvascular to be rendered at sub-wavelength resolution. Nevertheless, to obtain such superior spatial resolution, it is necessary to spend tens of seconds gathering numerous ultrasound (US) frames to accumulate MB events required, resulting in ULM imaging still suffering from trade-offs between imaging quality, data acquisition time and data processing speed. In this paper, we present a new deep learning (DL) framework combining multi-branch CNN and recursive Transformer, termed as ULM-MbCNRT, that is capable of reconstructing a super-resolution image directly from a temporal mean low-resolution image generated by averaging much fewer raw US frames, i.e., implement an ultrafast ULM imaging. To evaluate the performance of ULM-MbCNRT, a series of numerical simulations and in vivo experiments are carried out. Numerical simulation results indicate that ULM-MbCNRT achieves high-quality ULM imaging with ~10-fold reduction in data acquisition time and ~130-fold reduction in computation time compared to the previous DL method (e.g., the modified sub-pixel convolutional neural network, ULM-mSPCN). For the in vivo experiments, when comparing to the ULM-mSPCN, ULM-MbCNRT allows ~37-fold reduction in data acquisition time (~0.8 s) and ~2134-fold reduction in computation time (~0.87 s) without sacrificing spatial resolution. It implies that ultrafast ULM imaging holds promise for observing rapid biological activity in vivo, potentially improving the diagnosis and monitoring of clinical conditions.

PMID:38607709 | DOI:10.1109/TUFFC.2024.3388102

IEEE Trans Image Process. 2024 Apr 12;PP. doi: 10.1109/TIP.2024.3355818. Online ahead of print.

ABSTRACT

With recent deep learning based approaches showing promising results in removing noise from images, the best denoising performance has been reported in a supervised learning setup that requires a large set of paired noisy images and ground truth data for training. The strong data requirement can be mitigated by unsupervised learning techniques, however, accurate modelling of images or noise variances is still crucial for high-quality solutions. The learning problem is ill-posed for unknown noise distributions. This paper investigates the tasks of image denoising and noise variance estimation in a single, joint learning framework. To address the ill-posedness of the problem, we present deep variation prior (DVP), which states that the variation of a properly learnt denoiser with respect to the change of noise satisfies some smoothness properties, as a key criterion for good denoisers. Building upon DVP and under the assumption that the noise is zero mean and pixel-wise independent conditioned on the image, an unsupervised deep learning framework, that simultaneously learns a denoiser and estimates noise variances, is developed. Our method does not require any clean training images or an external step of noise estimation, and instead, approximates the minimum mean squared error denoisers using only a set of noisy images. With the two underlying tasks being considered in a single framework, we allow them to be optimised for each other. The experimental results show a denoising quality comparable to that of supervised learning and accurate noise variance estimates.

PMID:38607702 | DOI:10.1109/TIP.2024.3355818

Odontology. 2024 Apr 12. doi: 10.1007/s10266-024-00933-1. Online ahead of print.

ABSTRACT

The objectives of this study were to create a mutual conversion system between contrast-enhanced computed tomography (CECT) and non-CECT images using a cycle generative adversarial network (cycleGAN) for the internal jugular region. Image patches were cropped from CT images in 25 patients who underwent both CECT and non-CECT imaging. Using a cycleGAN, synthetic CECT and non-CECT images were generated from original non-CECT and CECT images, respectively. The peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) were calculated. Visual Turing tests were used to determine whether oral and maxillofacial radiologists could tell the difference between synthetic versus original images, and receiver operating characteristic (ROC) analyses were used to assess the radiologists' performances in discriminating lymph nodes from blood vessels. The PSNR of non-CECT images was higher than that of CECT images, while the SSIM was higher in CECT images. The Visual Turing test showed a higher perceptual quality in CECT images. The area under the ROC curve showed almost perfect performances in synthetic as well as original CECT images. In conclusion, synthetic CECT images created by cycleGAN appeared to have the potential to provide effective information in patients who could not receive contrast enhancement.

PMID:38607582 | DOI:10.1007/s10266-024-00933-1

EJNMMI Res. 2024 Apr 12;14(1):38. doi: 10.1186/s13550-023-01059-1.

ABSTRACT

BACKGROUND: The total-body positron emission tomography/computed tomography (PET/CT) system, with a long axial field of view, represents the state-of-the-art PET imaging technique. Recently, the total-body PET/CT system has been commercially available. The total-body PET/CT system enables high-resolution whole-body imaging, even under extreme conditions such as ultra-low dose, extremely fast imaging speed, delayed imaging more than 10 h after tracer injection, and total-body dynamic scan. The total-body PET/CT system provides a real-time picture of the tracers of all organs across the body, which not only helps to explain normal human physiological process, but also facilitates the comprehensive assessment of systemic diseases. In addition, the total-body PET/CT system may play critical roles in other medical fields, including cancer imaging, drug development and immunology.

MAIN BODY: Therefore, it is of significance to summarize the existing studies of the total-body PET/CT systems and point out its future direction. This review collected research literatures from the PubMed database since the advent of commercially available total-body PET/CT systems to the present, and was divided into the following sections: Firstly, a brief introduction to the total-body PET/CT system was presented, followed by a summary of the literature on the performance evaluation of the total-body PET/CT. Then, the research and clinical applications of the total-body PET/CT were discussed. Fourthly, deep learning studies based on total-body PET imaging was reviewed. At last, the shortcomings of existing research and future directions for the total-body PET/CT were discussed.

CONCLUSION: Due to its technical advantages, the total-body PET/CT system is bound to play a greater role in clinical practice in the future.

PMID:38607510 | DOI:10.1186/s13550-023-01059-1

Radiol Imaging Cancer. 2024 May;6(3):e230107. doi: 10.1148/rycan.230107.

ABSTRACT

Purpose To develop a custom deep convolutional neural network (CNN) for noninvasive prediction of breast cancer nodal metastasis. Materials and Methods This retrospective study included patients with newly diagnosed primary invasive breast cancer with known pathologic (pN) and clinical nodal (cN) status who underwent dynamic contrast-enhanced (DCE) breast MRI at the authors' institution between July 2013 and July 2016. Clinicopathologic data (age, estrogen receptor and human epidermal growth factor 2 status, Ki-67 index, and tumor grade) and cN and pN status were collected. A four-dimensional (4D) CNN model integrating temporal information from dynamic image sets was developed. The convolutional layers learned prognostic image features, which were combined with clinicopathologic measures to predict cN0 versus cN+ and pN0 versus pN+ disease. Performance was assessed with the area under the receiver operating characteristic curve (AUC), with fivefold nested cross-validation. Results Data from 350 female patients (mean age, 51.7 years ± 11.9 [SD]) were analyzed. AUC, sensitivity, and specificity values of the 4D hybrid model were 0.87 (95% CI: 0.83, 0.91), 89% (95% CI: 79%, 93%), and 76% (95% CI: 68%, 88%) for differentiating pN0 versus pN+ and 0.79 (95% CI: 0.76, 0.82), 80% (95% CI: 77%, 84%), and 62% (95% CI: 58%, 67%), respectively, for differentiating cN0 versus cN+. Conclusion The proposed deep learning model using tumor DCE MR images demonstrated high sensitivity in identifying breast cancer lymph node metastasis and shows promise for potential use as a clinical decision support tool. Keywords: MR Imaging, Breast, Breast Cancer, Breast MRI, Machine Learning, Metastasis, Prognostic Prediction Supplemental material is available for this article. Published under a CC BY 4.0 license.

PMID:38607282 | DOI:10.1148/rycan.230107

Tumori. 2024 Apr 12:3008916241231035. doi: 10.1177/03008916241231035. Online ahead of print.

ABSTRACT

Artificial intelligence (AI) applications in oncology are at the forefront of transforming healthcare during the Fourth Industrial Revolution, driven by the digital data explosion. This review provides an accessible introduction to the field of AI, presenting a concise yet structured overview of the foundations of AI, including expert systems, classical machine learning, and deep learning, along with their contextual application in clinical research and healthcare. We delve into the current applications of AI in oncology, with a particular focus on diagnostic imaging and pathology. Numerous AI tools have already received regulatory approval, and more are under active development, bringing clear benefits but not without challenges. We discuss the importance of data security, the need for transparent and interpretable models, and the ethical considerations that must guide AI development in healthcare. By providing a perspective on the opportunities and challenges, this review aims to inform and guide researchers, clinicians, and policymakers in the adoption of AI in oncology.

PMID:38606831 | DOI:10.1177/03008916241231035

Epilepsia. 2024 Apr 12. doi: 10.1111/epi.17984. Online ahead of print.

ABSTRACT

OBJECTIVE: Recently, a deep learning artificial intelligence (AI) model forecasted seizure risk using retrospective seizure diaries with higher accuracy than random forecasts. The present study sought to prospectively evaluate the same algorithm.

METHODS: We recruited a prospective cohort of 46 people with epilepsy; 25 completed sufficient data entry for analysis (median = 5 months). We used the same AI method as in our prior study. Group-level and individual-level Brier Skill Scores (BSSs) compared random forecasts and simple moving average forecasts to the AI.

RESULTS: The AI had an area under the receiver operating characteristic curve of .82. At the group level, the AI outperformed random forecasting (BSS = .53). At the individual level, AI outperformed random in 28% of cases. At the group and individual level, the moving average outperformed the AI. If pre-enrollment (nonverified) diaries (with presumed underreporting) were included, the AI significantly outperformed both comparators. Surveys showed most did not mind poor-quality LOW-RISK or HIGH-RISK forecasts, yet 91% wanted access to these forecasts.

SIGNIFICANCE: The previously developed AI forecasting tool did not outperform a very simple moving average forecasting in this prospective cohort, suggesting that the AI model should be replaced.

PMID:38606580 | DOI:10.1111/epi.17984

IEEE Open J Eng Med Biol. 2024 Apr 5;5:238-249. doi: 10.1109/OJEMB.2024.3377461. eCollection 2024.

ABSTRACT

Goal: Contractile response and calcium handling are central to understanding cardiac function and physiology, yet existing methods of analysis to quantify these metrics are often time-consuming, prone to mistakes, or require specialized equipment/license. We developed BeatProfiler, a suite of cardiac analysis tools designed to quantify contractile function, calcium handling, and force generation for multiple in vitro cardiac models and apply downstream machine learning methods for deep phenotyping and classification. Methods: We first validate BeatProfiler's accuracy, robustness, and speed by benchmarking against existing tools with a fixed dataset. We further confirm its ability to robustly characterize disease and dose-dependent drug response. We then demonstrate that the data acquired by our automatic acquisition pipeline can be further harnessed for machine learning (ML) analysis to phenotype a disease model of restrictive cardiomyopathy and profile cardioactive drug functional response. To accurately classify between these biological signals, we apply feature-based ML and deep learning models (temporal convolutional-bidirectional long short-term memory model or TCN-BiLSTM). Results: Benchmarking against existing tools revealed that BeatProfiler detected and analyzed contraction and calcium signals better than existing tools through improved sensitivity in low signal data, reduction in false positives, and analysis speed increase by 7 to 50-fold. Of signals accurately detected by published methods (PMs), BeatProfiler's extracted features showed high correlations to PMs, confirming that it is reliable and consistent with PMs. The features extracted by BeatProfiler classified restrictive cardiomyopathy cardiomyocytes from isogenic healthy controls with 98% accuracy and identified relax90 as a top distinguishing feature in congruence with previous findings. We also show that our TCN-BiLSTM model was able to classify drug-free control and 4 cardiac drugs with different mechanisms of action at 96% accuracy. We further apply Grad-CAM on our convolution-based models to identify signature regions of perturbations by these drugs in calcium signals. Conclusions: We anticipate that the capabilities of BeatProfiler will help advance in vitro studies in cardiac biology through rapid phenotyping, revealing mechanisms underlying cardiac health and disease, and enabling objective classification of cardiac disease and responses to drugs.

PMID:38606403 | PMC:PMC11008807 | DOI:10.1109/OJEMB.2024.3377461

IEEE Open J Eng Med Biol. 2024 Mar 12;5:226-237. doi: 10.1109/OJEMB.2024.3374966. eCollection 2024.

ABSTRACT

Recently, deep learning-based methods have emerged as the preferred approach for ultrasound data analysis. However, these methods often require large-scale annotated datasets for training deep models, which are not readily available in practical scenarios. Additionally, the presence of speckle noise and other imaging artifacts can introduce numerous hard examples for ultrasound data classification. In this paper, drawing inspiration from self-supervised learning techniques, we present a pre-training method based on mask modeling specifically designed for ultrasound data. Our study investigates three different mask modeling strategies: random masking, vertical masking, and horizontal masking. By employing these strategies, our pre-training approach aims to predict the masked portion of the ultrasound images. Notably, our method does not rely on externally labeled data, allowing us to extract representative features without the need for human annotation. Consequently, we can leverage unlabeled datasets for pre-training. Furthermore, to address the challenges posed by hard samples in ultrasound data, we propose a novel hard sample mining strategy. To evaluate the effectiveness of our proposed method, we conduct experiments on two datasets. The experimental results demonstrate that our approach outperforms other state-of-the-art methods in ultrasound image classification. This indicates the superiority of our pre-training method and its ability to extract discriminative features from ultrasound data, even in the presence of hard examples.

PMID:38606402 | PMC:PMC11008806 | DOI:10.1109/OJEMB.2024.3374966

IEEE Open J Eng Med Biol. 2024 Feb 20;5:216-225. doi: 10.1109/OJEMB.2024.3367243. eCollection 2024.

ABSTRACT

Goal: Cervical cancer is one of the most common cancers in women worldwide, ranking among the top four. Unfortunately, it is also the fourth leading cause of cancer-related deaths among women, particularly in developing countries where incidence and mortality rates are higher compared to developed nations. Colposcopy can aid in the early detection of cervical lesions, but its effectiveness is limited in areas with limited medical resources and a lack of specialized physicians. Consequently, many cases are diagnosed at later stages, putting patients at significant risk. Methods: This paper proposes an automated colposcopic image analysis framework to address these challenges. The framework aims to reduce the labor costs associated with cervical precancer screening in undeserved regions and assist doctors in diagnosing patients. The core of the framework is the MFEM-CIN hybrid model, which combines Convolutional Neural Networks (CNN) and Transformer to aggregate the correlation between local and global features. This combined analysis of local and global information is scientifically useful in clinical diagnosis. In the model, MSFE and MSFF are utilized to extract and fuse multi-scale semantics. This preserves important shallow feature information and allows it to interact with the deep feature, enriching the semantics to some extent. Conclusions: The experimental results demonstrate an accuracy rate of 89.2% in identifying cervical intraepithelial neoplasia while maintaining a lightweight model. This performance exceeds the average accuracy achieved by professional physicians, indicating promising potential for practical application. Utilizing automated colposcopic image analysis and the MFEM-CIN model, this research offers a practical solution to reduce the burden on healthcare providers and improve the efficiency and accuracy of cervical cancer diagnosis in resource-constrained areas.

PMID:38606400 | PMC:PMC11008799 | DOI:10.1109/OJEMB.2024.3367243

IEEE Open J Eng Med Biol. 2024 Feb 20;5:191-197. doi: 10.1109/OJEMB.2024.3367492. eCollection 2024.

ABSTRACT

Goal: To predict physician fixations specifically on ophthalmology optical coherence tomography (OCT) reports from eye tracking data using CNN based saliency prediction methods in order to aid in the education of ophthalmologists and ophthalmologists-in-training. Methods: Fifteen ophthalmologists were recruited to each examine 20 randomly selected OCT reports and evaluate the likelihood of glaucoma for each report on a scale of 0-100. Eye movements were collected using a Pupil Labs Core eye-tracker. Fixation heat maps were generated using fixation data. Results: A model trained with traditional saliency mapping resulted in a correlation coefficient (CC) value of 0.208, a Normalized Scanpath Saliency (NSS) value of 0.8172, a Kullback-Leibler (KLD) value of 2.573, and a Structural Similarity Index (SSIM) of 0.169. Conclusions: The TranSalNet model was able to predict fixations within certain regions of the OCT report with reasonable accuracy, but more data is needed to improve model accuracy. Future steps include increasing data collection, improving quality of data, and modifying the model architecture.

PMID:38606397 | PMC:PMC11008801 | DOI:10.1109/OJEMB.2024.3367492

IEEE J Transl Eng Health Med. 2024 Mar 21;12:401-412. doi: 10.1109/JTEHM.2024.3368106. eCollection 2024.

ABSTRACT

Osteoporosis is a prevalent chronic disease worldwide, particularly affecting the aging population. The gold standard diagnostic tool for osteoporosis is Dual-energy X-ray Absorptiometry (DXA). However, the expensive cost of the DXA machine and the need for skilled professionals to operate it restrict its accessibility to the general public. This paper builds upon previous research and proposes a novel approach for rapidly screening bone density. The method involves utilizing near-infrared light to capture local body information within the human body. Deep learning techniques are employed to analyze the obtained data and extract meaningful insights related to bone density. Our initial prediction, utilizing multi-linear regression, demonstrated a strong correlation (r = 0.98, p-value = 0.003**) with the measured Bone Mineral Density (BMD) obtained from Dual-energy X-ray Absorptiometry (DXA). This indicates a highly significant relationship between the predicted values and the actual BMD measurements. A deep learning-based algorithm is applied to analyze the underlying information further to predict bone density at the wrist, hip, and spine. The prediction of bone densities in the hip and spine holds significant importance due to their status as gold-standard sites for assessing an individual's bone density. Our prediction rate had an error margin below 10% for the wrist and below 20% for the hip and spine bone density.

PMID:38606393 | PMC:PMC11008809 | DOI:10.1109/JTEHM.2024.3368106

IEEE J Transl Eng Health Med. 2024 Feb 26;12:359-370. doi: 10.1109/JTEHM.2024.3369764. eCollection 2024.

ABSTRACT

Attention Deficit Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder commonly seen in childhood that leads to behavioural changes in social development and communication patterns, often continues into undiagnosed adulthood due to a global shortage of psychiatrists, resulting in delayed diagnoses with lasting consequences on individual's well-being and the societal impact. Recently, machine learning methodologies have been incorporated into healthcare systems to facilitate the diagnosis and enhance the potential prediction of treatment outcomes for mental health conditions. In ADHD detection, the previous research focused on utilizing functional magnetic resonance imaging (fMRI) or Electroencephalography (EEG) signals, which require costly equipment and trained personnel for data collection. In recent years, speech and text modalities have garnered increasing attention due to their cost-effectiveness and non-wearable sensing in data collection. In this research, conducted in collaboration with the Cumbria, Northumberland, Tyne and Wear NHS Foundation Trust, we gathered audio data from both ADHD patients and normal controls based on the clinically popular Diagnostic Interview for ADHD in adults (DIVA). Subsequently, we transformed the speech data into text modalities through the utilization of the Google Cloud Speech API. We extracted both acoustic and text features from the data, encompassing traditional acoustic features (e.g., MFCC), specialized feature sets (e.g., eGeMAPS), as well as deep-learned linguistic and semantic features derived from pre-trained deep learning models. These features are employed in conjunction with a support vector machine for ADHD classification, yielding promising outcomes in the utilization of audio and text data for effective adult ADHD screening. Clinical impact: This research introduces a transformative approach in ADHD diagnosis, employing speech and text analysis to facilitate early and more accessible detection, particularly beneficial in areas with limited psychiatric resources. Clinical and Translational Impact Statement: The successful application of machine learning techniques in analyzing audio and text data for ADHD screening represents a significant advancement in mental health diagnostics, paving the way for its integration into clinical settings and potentially improving patient outcomes on a broader scale.

PMID:38606391 | PMC:PMC11008805 | DOI:10.1109/JTEHM.2024.3369764

IEEE Winter Conf Appl Comput Vis. 2024 Jan;2024:7867-7876. doi: 10.1109/wacv57701.2024.00770. Epub 2024 Apr 9.

ABSTRACT

Age is one of the major known risk factors for Alzheimer's Disease (AD). Detecting AD early is crucial for effective treatment and preventing irreversible brain damage. Brain age, a measure derived from brain imaging reflecting structural changes due to aging, may have the potential to identify AD onset, assess disease risk, and plan targeted interventions. Deep learning-based regression techniques to predict brain age from magnetic resonance imaging (MRI) scans have shown great accuracy recently. However, these methods are subject to an inherent regression to the mean effect, which causes a systematic bias resulting in an overestimation of brain age in young subjects and underestimation in old subjects. This weakens the reliability of predicted brain age as a valid biomarker for downstream clinical applications. Here, we reformulate the brain age prediction task from regression to classification to address the issue of systematic bias. Recognizing the importance of preserving ordinal information from ages to understand aging trajectory and monitor aging longitudinally, we propose a novel ORdinal Distance Encoded Regularization (ORDER) loss that incorporates the order of age labels, enhancing the model's ability to capture age-related patterns. Extensive experiments and ablation studies demonstrate that this framework reduces systematic bias, outperforms state-of-art methods by statistically significant margins, and can better capture subtle differences between clinical groups in an independent AD dataset. Our implementation is publicly available at https://github.com/jaygshah/Robust-Brain-Age-Prediction.

PMID:38606366 | PMC:PMC11008505 | DOI:10.1109/wacv57701.2024.00770

Front Neurosci. 2024 Mar 28;18:1309594. doi: 10.3389/fnins.2024.1309594. eCollection 2024.

ABSTRACT

INTRODUCTION: Within the development of brain-computer interface (BCI) systems, it is crucial to consider the impact of brain network dynamics and neural signal transmission mechanisms on electroencephalogram-based motor imagery (MI-EEG) tasks. However, conventional deep learning (DL) methods cannot reflect the topological relationship among electrodes, thereby hindering the effective decoding of brain activity.

METHODS: Inspired by the concept of brain neuronal forward-forward (F-F) mechanism, a novel DL framework based on Graph Neural Network combined forward-forward mechanism (F-FGCN) is presented. F-FGCN framework aims to enhance EEG signal decoding performance by applying functional topological relationships and signal propagation mechanism. The fusion process involves converting the multi-channel EEG into a sequence of signals and constructing a network grounded on the Pearson correlation coeffcient, effectively representing the associations between channels. Our model initially pre-trains the Graph Convolutional Network (GCN), and fine-tunes the output layer to obtain the feature vector. Moreover, the F-F model is used for advanced feature extraction and classification.

RESULTS AND DISCUSSION: Achievement of F-FGCN is assessed on the PhysioNet dataset for a four-class categorization, compared with various classical and state-of-the-art models. The learned features of the F-FGCN substantially amplify the performance of downstream classifiers, achieving the highest accuracy of 96.11% and 82.37% at the subject and group levels, respectively. Experimental results affirm the potency of FFGCN in enhancing EEG decoding performance, thus paving the way for BCI applications.

PMID:38606308 | PMC:PMC11008472 | DOI:10.3389/fnins.2024.1309594

Front Neurosci. 2024 Mar 28;18:1335422. doi: 10.3389/fnins.2024.1335422. eCollection 2024.

ABSTRACT

Neuromorphic processors promise low-latency and energy-efficient processing by adopting novel brain-inspired design methodologies. Yet, current neuromorphic solutions still struggle to rival conventional deep learning accelerators' performance and area efficiency in practical applications. Event-driven data-flow processing and near/in-memory computing are the two dominant design trends of neuromorphic processors. However, there remain challenges in reducing the overhead of event-driven processing and increasing the mapping efficiency of near/in-memory computing, which directly impacts the performance and area efficiency. In this work, we discuss these challenges and present our exploration of optimizing event-based neural network inference on SENECA, a scalable and flexible neuromorphic architecture. To address the overhead of event-driven processing, we perform comprehensive design space exploration and propose spike-grouping to reduce the total energy and latency. Furthermore, we introduce the event-driven depth-first convolution to increase area efficiency and latency in convolutional neural networks (CNNs) on the neuromorphic processor. We benchmarked our optimized solution on keyword spotting, sensor fusion, digit recognition and high resolution object detection tasks. Compared with other state-of-the-art large-scale neuromorphic processors, our proposed optimizations result in a 6× to 300× improvement in energy efficiency, a 3× to 15× improvement in latency, and a 3× to 100× improvement in area efficiency. Our optimizations for event-based neural networks can be potentially generalized to a wide range of event-based neuromorphic processors.

PMID:38606307 | PMC:PMC11007209 | DOI:10.3389/fnins.2024.1335422

Genome Med. 2024 Apr 11;16(1):55. doi: 10.1186/s13073-024-01327-2.

ABSTRACT

BACKGROUND: Most primary Triple Negative Breast Cancers (TNBCs) show amplification of the Epidermal Growth Factor Receptor (EGFR) gene, leading to increased protein expression. However, unlike other EGFR-driven cancers, targeting this receptor in TNBC yields inconsistent therapeutic responses.

METHODS: To elucidate the underlying mechanisms of this variability, we employ cellular barcoding and single-cell transcriptomics to reconstruct the subclonal dynamics of EGFR-amplified TNBC cells in response to afatinib, a tyrosine kinase inhibitor (TKI) that irreversibly inhibits EGFR.

RESULTS: Integrated lineage tracing analysis revealed a rare pre-existing subpopulation of cells with distinct biological signature, including elevated expression levels of Insulin-Like Growth Factor Binding Protein 2 (IGFBP2). We show that IGFBP2 overexpression is sufficient to render TNBC cells tolerant to afatinib treatment by activating the compensatory insulin-like growth factor I receptor (IGF1-R) signalling pathway. Finally, based on reconstructed mechanisms of resistance, we employ deep learning techniques to predict the afatinib sensitivity of TNBC cells.

CONCLUSIONS: Our strategy proved effective in reconstructing the complex signalling network driving EGFR-targeted therapy resistance, offering new insights for the development of individualized treatment strategies in TNBC.

PMID:38605363 | DOI:10.1186/s13073-024-01327-2

Commun Med (Lond). 2024 Apr 11;4(1):72. doi: 10.1038/s43856-024-00496-w.

ABSTRACT

BACKGROUND: Sensory changes due to aging or disease can impact brain tissue. This study aims to investigate the link between glaucoma, a leading cause of blindness, and alterations in brain connections.

METHODS: We analyzed diffusion MRI measurements of white matter tissue in a large group, consisting of 905 glaucoma patients (aged 49-80) and 5292 healthy individuals (aged 45-80) from the UK Biobank. Confounds due to group differences were mitigated by matching a sub-sample of controls to glaucoma subjects. We compared classification of glaucoma using convolutional neural networks (CNNs) focusing on the optic radiations, which are the primary visual connection to the cortex, against those analyzing non-visual brain connections. As a control, we evaluated the performance of regularized linear regression models.

RESULTS: We showed that CNNs using information from the optic radiations exhibited higher accuracy in classifying subjects with glaucoma when contrasted with CNNs relying on information from non-visual brain connections. Regularized linear regression models were also tested, and showed significantly weaker classification performance. Additionally, the CNN was unable to generalize to the classification of age-group or of age-related macular degeneration.

CONCLUSIONS: Our findings indicate a distinct and potentially non-linear signature of glaucoma in the tissue properties of optic radiations. This study enhances our understanding of how glaucoma affects brain tissue and opens avenues for further research into how diseases that affect sensory input may also affect brain aging.

PMID:38605245 | DOI:10.1038/s43856-024-00496-w

Sci Data. 2024 Apr 11;11(1):365. doi: 10.1038/s41597-024-03182-7.

ABSTRACT

Optical coherence tomography (OCT) is a non-invasive imaging technique with extensive clinical applications in ophthalmology. OCT enables the visualization of the retinal layers, playing a vital role in the early detection and monitoring of retinal diseases. OCT uses the principle of light wave interference to create detailed images of the retinal microstructures, making it a valuable tool for diagnosing ocular conditions. This work presents an open-access OCT dataset (OCTDL) comprising over 2000 OCT images labeled according to disease group and retinal pathology. The dataset consists of OCT records of patients with Age-related Macular Degeneration (AMD), Diabetic Macular Edema (DME), Epiretinal Membrane (ERM), Retinal Artery Occlusion (RAO), Retinal Vein Occlusion (RVO), and Vitreomacular Interface Disease (VID). The images were acquired with an Optovue Avanti RTVue XR using raster scanning protocols with dynamic scan length and image resolution. Each retinal b-scan was acquired by centering on the fovea and interpreted and cataloged by an experienced retinal specialist. In this work, we applied Deep Learning classification techniques to this new open-access dataset.

PMID:38605088 | DOI:10.1038/s41597-024-03182-7