Encoder-decoder Convolutional Neural Network Research Articles

DehazeModel: Enhancing Image Clarity with an Encoder-Decoder CNN Approach

We introduce DehazeModel, a Convolutional Neural Network (CNN) tailored for the purpose of enhancing single images by removing haze. DehazeModel comprises several components, including pre-processing, image dehazing, and post-processing modules. The pre-processing module, which is trainable, generates enhanced inputs featuring a wider array of characteristics compared to manually chosen pre-processing methods. The Image Dehazing module employs a novel encoder-decoder framework, overcoming common issues found in traditional multi-scale methods. Moreover, the post-processing module aids in minimizing artifacts in the resultant output. Experimental findings indicate that DehazeModel surpasses existing state-of-the-art techniques on the RESIDE dataset. Our experiments, conducted on both indoor and outdoor images, illustrate the robustness of our approach across various scenarios, independent of atmospheric scattering effects.

Automated detection and counting of broiler behaviors using a video recognition system

Stretching and preening behaviors in broilers are linked to their comfort and overall welfare. These behaviors signal that broilers engage in natural behaviors and physical activity and that the environment is conducive to their needs. By examining the frequency of stretching and preening, valuable information can be gained regarding broilers' physical health and environmental circumstances. This study attempted to quantify stretching and preening among individual broilers in commercial farms, aiming to fill a gap in the literature on this subject. The proposed video recognition-based pipeline consisted of four main components: 1) identification of broilers using an encoder-decoder convolutional neural network, 2) tracking the identified broilers using the Euclidean distance algorithm, 3) utilizing a 3D convolutional network to recognize stretching and preening actions, and 4) detecting the heads of broilers using the YOLO model. Three separate filtering mechanisms, which involve assessing the broiler's movement, its body size, and the alignment of the head with the body, were implemented to improve the performance of the pipeline. To attain the goals of this research, 2500 labeled images and 1500 labeled videos obtained from a commercial broiler farm were used to train the CNN-based models. After assessing the proposed algorithm with eight 15-minute videos, the average per-class of its performance parameters − accuracy, precision, sensitivity, specificity, and AUC − was found to be 96.70%, 88.10%, 89.96%, 94.92%, and 92.44%, respectively. These findings support the feasibility of implementing the proposed algorithm as an automated tool in poultry farms.

ConTEXTual Net: A Multimodal Vision-Language Model for Segmentation of Pneumothorax.

Radiology narrative reports often describe characteristics of a patient's disease, including its location, size, and shape. Motivated by the recent success of multimodal learning, we hypothesized that this descriptive text could guide medical image analysis algorithms. We proposed a novel vision-language model, ConTEXTual Net, for the task of pneumothorax segmentation on chest radiographs. ConTEXTual Net extracts language features from physician-generated free-form radiology reports using a pre-trained language model. We then introduced cross-attention between the language features and the intermediate embeddings of an encoder-decoder convolutional neural network to enable language guidance for image analysis. ConTEXTual Net was trained on the CANDID-PTX dataset consisting of 3196 positive cases of pneumothorax with segmentation annotations from 6 different physicians as well as clinical radiology reports. Using cross-validation, ConTEXTual Net achieved a Dice score of 0.716±0.016, which was similar to the degree of inter-reader variability (0.712±0.044) computed on a subset of the data. It outperformed vision-only models (Swin UNETR: 0.670±0.015, ResNet50 U-Net: 0.677±0.015, GLoRIA: 0.686±0.014, and nnUNet 0.694±0.016) and a competing vision-language model (LAVT: 0.706±0.009). Ablation studies confirmed that it was the text information that led to the performance gains. Additionally, we show that certain augmentation methods degraded ConTEXTual Net's segmentation performance by breaking the image-text concordance. We also evaluated the effects of using different language models and activation functions in the cross-attention module, highlighting the efficacy of our chosen architectural design.

Mild Traumatic Brain Injury andCareer Stage Associate with Visible Perivascular Spaces in Special Operations Forces Soldiers.

Mild traumatic brain injury (mTBI) and occupational blast exposure in military Service Members may lead to impaired brain waste clearance which increases neurological disease risk. Perivascular spaces (PVS) are a key part of the glymphatic system which supports brain waste clearance, preferentially during sleep. Visible PVS on clinical magnetic resonance imaging have been previously observed in patients with neurodegenerative diseases and animal neurotrauma models. The purpose of this study was to determine associations between PVS morphological characteristics, military career stage, and mTBI history in Special Operations Forces (SOF)Soldiers. Participants underwent T2-weighed neuroimaging to capture three-dimensional whole brain volumes. Segmentation was performed using a previously validated, multi-scale deep convolutional encoder-decoder neural network. Only PVS clusters within the white matter mask were quantified for analyses. Due to non-normal PVS metric distribution, non-parametric Mann-Whitney U tests were used to determine group differences in PVS outcomes. In total, 223 healthy SOF combat Soldiers (age = 33.1 ± 4.3yrs) were included, 217 reported career stage. Soldiers with mTBI history had greater PVS number (z = 2.51, P = 0.013) and PVS volume (z = 2.42, P = 0.016). In-career SOF combat Soldiers had greater PVS number (z = 2.56, P = 0.01) and PVS volume (z = 2.28, P = 0.02) compared to a baseline cohort. Mild TBI history is associated with increased PVS burden in SOF combat Soldiers that are clinically recovered from mTBI. This may indicate ongoing physiological changes that could lead to impaired waste clearance via the glymphatic system. Future studies should determine if PVS number and volume are meaningful neurobiological outcomes for neurodegenerative disease risk and if clinical interventions such as improving sleep can reduce PVS burden.

Characterization and depth detection of internal delamination defects in CFRP based on line laser scanning infrared thermography

In this article, the shape characterization and depth detection of delamination defects in carbon fiber reinforced polymer (CFRP) sandwich are achieved by detecting delamination defects through line laser infrared thermography. In terms of defect shape characterization, the temperature matrix difference imaging method is proposed to characterize the shape of internal defects, and the defect shape distortion is corrected by the principal component analysis algorithm to optimize the effect of defect shape characterization. In terms of defect depth detection, this study constructs a dataset of cooling curves for defects of different depths and builds an encoder–decoder convolutional neural network structure based on the attention mechanism, which is centered on a streamlined encoder–decoder structure as well as a highly integrated attention gate, which is capable of accelerating the extraction of features and reducing the number of iterations of the algorithm. The results demonstrate that we achieved an average classification accuracy of 88.9% for defects with a depth gradient difference of 0.2 mm, attaining the highest classification accuracy of 95% for a single defect. Furthermore, our approach outperformed several classical classifiers by achieving superior accuracy with fewer iterations. The experimental results demonstrate that the line laser scanning inspection method is effective in the task of detecting internal defects in CFRP materials.

Underwater Acoustic Signal Noise Reduction Based on a Fully Convolutional Encoder-Decoder Neural Network

Underwater Acoustic Signal Noise Reduction Based on a Fully Convolutional Encoder-Decoder Neural Network

Improved segmentation of basal ganglia from MR images using convolutional neural network with crossover-typed skip connection.

Accurate and automatic segmentation of basal ganglia from magnetic resonance (MR) images is important for diagnosis and treatment of various brain disorders. However, the basal ganglia segmentation is a challenging task because of the class imbalance and the unclear boundaries among basal ganglia anatomical structures. Thus, we aim to present an encoder-decoder convolutional neural network (CNN)-based method for improved segmentation of basal ganglia by focusing on skip connections that determine the segmentation performance of encoder-decoder CNNs. We also aim to reveal the effect of skip connections on the segmentation of basal ganglia with unclear boundaries. We used the encoder-decoder CNNs with the following five patterns of skip connections: without skip connection, with full-resolution horizontal skip connection, with horizontal skip connections, with vertical skip connections, and with crossover-typed skip connections (the proposed method). We compared and evaluated the performance of the CNNs in the experiment of basal ganglia segmentation using T1-weighted MR brain images of 79 patients. The experimental results showed that the skip connections at each scale level help CNNs to acquire multi-scale image features, the vertical skip connections contribute on acquiring finer image features for segmentation of smaller anatomical structures with more blurred boundaries, and the crossover-typed skip connections, a combination of horizontal and vertical skip connections, provided better segmentation accuracy. This paper investigated the effect of skip connections on the basal ganglia segmentation and revealed the crossover-typed skip connections might be effective for improving the segmentation of basal ganglia with the class imbalance and the unclear boundaries.

MVCT image enhancement using reference-based encoder–decoder convolutional neural network

Daily MVCT (Megavoltage Computed Tomography) in TomoTherapy plays a crucial role in patients’ setup and dose reconstruction. However, MVCT images suffer from high noise and low tissue contrast due to the limited number of X-ray photons and low detector quantum efficiency (DQE). In this study, we propose an approach to enhance MVCT images using the KVCT (Kilovoltage Computed Tomography) image of the same patient as an auxiliary reference image. Specifically, we propose a reference-based encoder–decoder convolutional neural network (RefED-CNN) by incorporating a feature extraction and alignment (FEA) module to introduce the features of the reference image as side information into the MVCT image enhancement process. The FEA module automatically searches and aligns relevant features between the reference image and the noisy image, and transfers the relevant texture from reference KVCT images to the target MVCT image. Evaluations conducted on both phantom and real patient data show that our method outperforms other denoising methods by effectively reducing noise and preserving intricate structural details.

ESP-UNet: Encoder-Decoder Convolutional Neural Network with Edge-Enhanced Features for Liver Segmentation

ESP-UNet: Encoder-Decoder Convolutional Neural Network with Edge-Enhanced Features for Liver Segmentation

Pavement crack detection through a deep-learned asymmetric encoder-decoder convolutional neural network

ABSTRACT Crack detection on roads’ surfaces is an important issue in pavement management, as it provides an indication of the quality of the road and its deterioration over time. Pavement cracks are one of the most common types of damage observed on roads, and they can be seen visually. Despite the fact that it does not provide immediate resolution to the issue, understanding the extent of crack damage is essential for the upkeep of roads. This paper presents a novel approach to automatically detecting pavement cracks using the orthoimage generated by a consumer-grade photogrammetric Unmanned Aerial Vehicle (UAV) and a deep learning algorithm. We used an autoencoder Convolutional Neural Network (CNN) to train a dataset full of challenging factors such as road lines and marks, oil and colour spots, and water stains. The model was tested on a dataset of RGB patches of different patterns of cracks and achieved an overall accuracy (OA) and F1 score of about 0.98. The results demonstrate the effectiveness of the proposed method in accurately detecting pavement cracks in challenging real-world conditions. This approach provides an efficient and cost-effective solution for pavement crack detection, that can be used for measuring the road's quality and monitoring it.

Atmospheric correction of vegetation reflectance with simulation-trained deep learning for ground-based hyperspectral remote sensing

BackgroundVegetation spectral reflectance obtained with hyperspectral imaging (HSI) offer non-invasive means for the non-destructive study of their physiological status. The light intensity at visible and near-infrared wavelengths (VNIR, 0.4–1.0µm) captured by the sensor are composed of mixtures of spectral components that include the vegetation reflectance, atmospheric attenuation, top-of-atmosphere solar irradiance, and sensor artifacts. Common methods for the extraction of spectral reflectance from the at-sensor spectral radiance offer a trade-off between explicit knowledge of atmospheric conditions and concentrations, computational efficiency, and prediction accuracy, and are generally geared towards nadir pointing platforms. Therefore, a method is needed for the accurate extraction of vegetation reflectance from spectral radiance captured by ground-based remote sensors with a side-facing orientation towards the target, and a lack of knowledge of the atmospheric parameters.ResultsWe propose a framework for obtaining the vegetation spectral reflectance from at-sensor spectral radiance, which relies on a time-dependent Encoder-Decoder Convolutional Neural Network trained and tested using simulated spectra generated from radiative transfer modeling. Simulated at-sensor spectral radiance are produced from combining 1440 unique simulated solar angles and atmospheric absorption profiles, and 1000 different spectral reflectance curves of vegetation with various health indicator values, together with sensor artifacts. Creating an ensemble of 10 models, each trained and tested on a separate 10% of the dataset, results in the prediction of the vegetation spectral reflectance with a testing r2 of 98.1% (±0.4). This method produces consistently high performance with accuracies >90% for spectra with resolutions as low as 40 channels in VNIR each with 40 nm full width at half maximum (FWHM) and greater, and remains viable with accuracies >80% down to a resolution of 10 channels with 60 nm FWHM. When applied to real sensor obtained spectral radiance data, the predicted spectral reflectance curves showed general agreement and consistency with those corrected by the Compound Ratio method.ConclusionsWe propose a method that allows for the accurate estimation of the vegetation spectral reflectance from ground-based HSI platforms with sufficient spectral resolution. It is capable of extracting the vegetation spectral reflectance at high accuracy in the absence of knowledge of the exact atmospheric compositions and conditions at time of capture, and the lack of available sensor-measured spectral radiance and their true ground-truth spectral reflectance profiles.

IS-Net: Automatic Ischemic Stroke Lesion Segmentation on CT Images

Ischemic stroke is an acute cerebral vascular disease and makes up about 80% of all stroke cases. Non-contrast computed tomography (NCCT) is a widely applied imaging technique for ischemic stroke assessment. However, it is challenging to identify ischemic lesion on NCCT images due to its high variability in location, contrast, and geometry. In this work, we propose IS-Net, an encoder-decoder convolutional neural network for automatic ischemic stroke lesion segmentation on NCCT images. The proposed IS-Net takes hierarchical network as backbone while pyramid feature aggregation (PFA) module is designed to aggregate features from multi stages of backbone, and reasonable feature fusion strategy is considered in PFA to enhance multi-level propagation. To fully mine the boundary cues, edge constraint scheme is introduced by deep supervision which broadcasts low-level features to different modules. In addition, to overcome the limitation of fixed geometric structure of convolution for multi-range dependency exploitation, non-local parallel decoder is introduced with deformable convolution and self-attention. The proposed IS-Net is evaluated on manually labeled follow-up NCCT dataset composed of 1,004 cases (totally 9,020 images). The proposed IS-Net is compared with the state-of-the-art segmentation models and illustrates the highest score on segmentation criteria and sensitivity.

Detection of Sargassum from Sentinel Satellite Sensors Using Deep Learning Approach

Since 2011, the proliferation of brown macro-algae of the genus Sargassum has considerably increased in the North Tropical Atlantic Sea, all the way from the Gulf of Guinea to the Caribbean Sea and the Gulf of Mexico. The large amount of Sargassum aggregations in that area cause major beaching events, which have a significant impact on the local economy and the environment and are starting to present a real threat to public health. In such a context, it is crucial to collect spatial and temporal data of Sargassum aggregations to understand their dynamics and predict stranding. Lately, indexes based on satellite imagery such as the Maximum Chlorophyll Index (MCI) or the Alternative Floating Algae Index (AFAI), have been developed and used to detect these Sargassum aggregations. However, their accuracy is questionable as they tend to detect various non-Sargassum features. To overcome false positive detection biases encountered by the index-thresholding methods, we developed two new deep learning models specific for Sargassum detection based on an encoder–decoder convolutional neural network (CNN). One was tuned to spectral bands from the multispectral instrument (MSI) onboard Sentinel-2 satellites and the other to the Ocean and Land Colour Instrument (OLCI) onboard Sentinel-3 satellites. This specific new approach outperformed previous generalist deep learning models, such as ErisNet, UNet, and SegNet, in the detection of Sargassum from satellite images with the same training, with an F1-score of 0.88 using MSI images, and 0.76 using OLCI images. Indeed, the proposed CNN considered neighbor pixels, unlike ErisNet, and had fewer reduction levels than UNet and SegNet, allowing filiform objects such as Sargassum aggregations to be detected. Using both spectral and spatial features, it also yielded a better detection performance compared to algal index-based techniques. The CNN method proposed here recognizes new small aggregations that were previously undetected, provides more complete structures, and has a lower false-positive detection rate.

A deep learning based dual encoder–decoder framework for anatomical structure segmentation in chest X-ray images

Automated multi-organ segmentation plays an essential part in the computer-aided diagnostic (CAD) of chest X-ray fluoroscopy. However, developing a CAD system for the anatomical structure segmentation remains challenging due to several indistinct structures, variations in the anatomical structure shape among different individuals, the presence of medical tools, such as pacemakers and catheters, and various artifacts in the chest radiographic images. In this paper, we propose a robust deep learning segmentation framework for the anatomical structure in chest radiographs that utilizes a dual encoder–decoder convolutional neural network (CNN). The first network in the dual encoder–decoder structure effectively utilizes a pre-trained VGG19 as an encoder for the segmentation task. The pre-trained encoder output is fed into the squeeze-and-excitation (SE) to boost the network’s representation power, which enables it to perform dynamic channel-wise feature calibrations. The calibrated features are efficiently passed into the first decoder to generate the mask. We integrated the generated mask with the input image and passed it through a second encoder–decoder network with the recurrent residual blocks and an attention the gate module to capture the additional contextual features and improve the segmentation of the smaller regions. Three public chest X-ray datasets are used to evaluate the proposed method for multi-organs segmentation, such as the heart, lungs, and clavicles, and single-organ segmentation, which include only lungs. The results from the experiment show that our proposed technique outperformed the existing multi-class and single-class segmentation methods.

An Improved Encoder-Decoder CNN with Region-Based Filtering for Vibrant Colorization

Colorization is the practice of adding appropriate chromatic values to monochrome photographs or videos. A real-valued luminance image can be mapped to a three-dimensional color image. However, it is a severely ill-defined problem and not has a single solution. In this paper, an encoder-decoder Convolutional Neural Network (CNN) model is used for colorizing gray images where the encoder is a Densely Connected Convolutional Network (DenseNet) and the decoder is a conventional CNN. The DenseNet extracts image features from gray images and the conventional CNN outputs a * b * color channels. Due to a large number of desaturated color components compared to saturated color components in the training images, the saturated color components have a strong tendency towards desaturated color components in the predicted a * b * channel. To solve the problems, we rebalance the predicted a * b * color channel by smoothing every subregion individually using the average filter. 2 stage k-means clustering technique is applied to divide the subregions. Then we apply Gamma transformation in the entire a * b * channel to saturate the image. We compare our proposed method with several existing methods. From the experimental results, we see that our proposed method has made some notable improvements over the existing methods and color representation of gray-scale images by our proposed method is more plausible to visualize. Additionally, our suggested approach beats other approaches in terms of Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM) and Histogram.

Direct field-to-pattern monolithic design of holographic metasurface via residual encoder-decoder convolutional neural network

Complex-amplitude holographic metasurfaces (CAHMs) with the flexibility in modulating phase and amplitude profiles have been used to manipulate the propagation of wavefront with an unprecedented level, leading to higher image-reconstruction quality compared with their natural counterparts. However, prevailing design methods of CAHMs are based on Huygens-Fresnel theory, meta-atom optimization, numerical simulation and experimental verification, which results in a consumption of computing resources. Here, we applied residual encoder-decoder convolutional neural network to directly map the electric field distributions and input images for monolithic metasurface design. A pretrained network is firstly trained by the electric field distributions calculated by diffraction theory, which is subsequently migrated as transfer learning framework to map the simulated electric field distributions and input images. The training results show that the normalized mean pixel error is about 3% on dataset. As verification, the metasurface prototypes are fabricated, simulated and measured. The reconstructed electric field of reverse-engineered metasurface exhibits high similarity to the target electric field, which demonstrates the effectiveness of our design. Encouragingly, this work provides a monolithic field-to-pattern design method for CAHMs, which paves a new route for the direct reconstruction of metasurfaces.

6D Phase space diagnostics based on adaptively tuned physics-informed generative convolutional neural networks

A physics-informed generative convolutional neural network (CNN)-based 6D phase space diagnostic is presented which generates all 15 unique 2D projections (x, y), (x, y′),...,(z, E) of a charged particle beam’s 6D phase space (x, y, z, x′, y′, E). The CNN is trained by supervised learning over a wide range of input beam distributions, accelerator parameters, and the associated 6D beam phase spaces at multiple accelerator locations. The CNN is applied in an un-supervised adaptive manner without knowledge of the input beam distribution or accelerator parameters and is robust to their unknown time variation. Adaptive feedback automatically tunes the low-dimensional latent space of the encoder-decoder CNN to predict the 6D phase space based only on 2D (z, E) longitudinal phase space measurements from a device such as a transverse deflecting RF cavity (TCAV). This method has the potential to provide diagnostics beyond the existing state of the art at many accelerator facilities. Studies are presented for two very different accelerators: the 5-meter-long ultra-fast electron diffraction (UED) HiRES compact accelerator at LBNL and the kilometer long plasma wakefield accelerator FACET-II at SLAC.

Robust automatic hexahedral cartilage meshing framework enables population-based computational studies of the knee.

Osteoarthritis of the knee is increasingly prevalent as our population ages, representing an increasing financial burden, and severely impacting quality of life. The invasiveness of in vivo procedures and the high cost of cadaveric studies has left computational tools uniquely suited to study knee biomechanics. Developments in deep learning have great potential for efficiently generating large-scale datasets to enable researchers to perform population-sized investigations, but the time and effort associated with producing robust hexahedral meshes has been a limiting factor in expanding finite element studies to encompass a population. Here we developed a fully automated pipeline capable of taking magnetic resonance knee images and producing a working finite element simulation. We trained an encoder-decoder convolutional neural network to perform semantic image segmentation on the Imorphics dataset provided through the Osteoarthritis Initiative. The Imorphics dataset contained 176 image sequences with varying levels of cartilage degradation. Starting from an open-source swept-extrusion meshing algorithm, we further developed this algorithm until it could produce high quality meshes for every sequence and we applied a template-mapping procedure to automatically place soft-tissue attachment points. The meshing algorithm produced simulation-ready meshes for all 176 sequences, regardless of the use of provided (manually reconstructed) or predicted (automatically generated) segmentation labels. The average time to mesh all bones and cartilage tissues was less than 2min per knee on an AMD Ryzen 5600X processor, using a parallel pool of three workers for bone meshing, followed by a pool of four workers meshing the four cartilage tissues. Of the 176 sequences with provided segmentation labels, 86% of the resulting meshes completed a simulated flexion-extension activity. We used a reserved testing dataset of 28 sequences unseen during network training to produce simulations derived from predicted labels. We compared tibiofemoral contact mechanics between manual and automated reconstructions for the 24 pairs of successful finite element simulations from this set, resulting in mean root-mean-squared differences under 20% of their respective min-max norms. In combination with further advancements in deep learning, this framework represents a feasible pipeline to produce population sized finite element studies of the natural knee from subject-specific models.

Noise Suppression of Computed Tomography (CT) Images Using Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)

In this study, an in-house residual encoder-decoder convolutional neural network (RED-CNN)-based algorithm was composed and trained using images of cylindrical polymethyl-methacrylate (PMMA) phantom with a diameter of 26 cm at different simulated noise levels. The model was tested on 21 × 26 cm elliptical PMMA computed tomography (CT) phantom images with simulated noise to evaluate its denoising capability using signal to noise ratio (SNR), comparative peak signal-to-noise ratio (cPSNR), structural similarity (SSIM) index, modulation transfer function frequencies (MTF 10 %) and noise power spectra (NPS) values as parameters. Evaluation of a possible decrease of image quality was also performed by testing the model using homogenous water phantom and wire phantom images acquired using different mAs values. Results show that the model was able to consistently increase SNR, cPSNR, SSIM values, and decrease the integral noise power spectra (NPS). However, the noise level on either training or testing data affects the model’s final denoising performance. The lower noise level on testing data images tends to result in over-smoothed images, as indicated by the shift of the NPS curves. In contrast, higher simulated noise level tends to result in less satisfactory denoising performance, as indicated by lower SNR, cPSNR, and SSIM values. Meanwhile, the higher noise level on training data images tends to produce denoised images with reduced sharpness, as indicated by the decrease of the MTF 10 % values. Further studies are required to better understand the character of RED-CNN for CT noise suppression regarding the optimum parameters for best results.

Streamlined single-molecule RNA-FISH of core clock mRNAs in clock neurons in whole mount Drosophila brains.

Circadian clocks are ∼24-h timekeepers that control rhythms in almost all aspects of our behavior and physiology. While it is well known that subcellular localization of core clock proteins plays a critical role in circadian regulation, very little is known about the spatiotemporal organization of core clock mRNAs and its role in generating ∼24-h circadian rhythms. Here we describe a streamlined single molecule Fluorescence In Situ Hybridization (smFISH) protocol and a fully automated analysis pipeline to precisely quantify the number and subcellular location of mRNAs of Clock, a core circadian transcription factor, in individual clock neurons in whole mount Drosophila adult brains. Specifically, we used ∼48 fluorescent oligonucleotide probes that can bind to an individual Clock mRNA molecule, which can then be detected as a diffraction-limited spot. Further, we developed a machine learning-based approach for 3-D cell segmentation, based on a pretrained encoder-decoder convolutional neural network, to automatically identify the cytoplasm and nuclei of clock neurons. We combined our segmentation model with a spot counting algorithm to detect Clock mRNA spots in individual clock neurons. Our results demonstrate that the number of Clock mRNA molecules cycle in large ventral lateral clock neurons (lLNvs) with peak levels at ZT4 (4h after lights are turned on) with ∼80 molecules/neuron and trough levels at ZT16 with ∼30 molecules/neuron. Our streamlined smFISH protocol and deep learning-based analysis pipeline can be employed to quantify the number and subcellular location of any mRNA in individual clock neurons in Drosophila brains. Further, this method can open mechanistic and functional studies into how spatiotemporal localization of clock mRNAs affect circadian rhythms.