AUTOMATED MORPHOLOGICAL ANALYSIS OF NEURONS IN FLUORESCENCE MICROSCOPY USING YOLOV8

BackgroundManual assessment and evaluation of fluorescent micrograph cell experiments is time-consuming and tedious. Automated segmentation pipelines can ensure efficient and reproducible evaluation and analysis with constant high quality for all images of an experiment. Such cell segmentation approaches are usually validated and rated in comparison to manually annotated micrographs. Nevertheless, manual annotations are prone to errors and display inter- and intra-observer variability which influence the validation results of automated cell segmentation pipelines.ResultsWe present a new approach to simulate fluorescent cell micrographs that provides an objective ground truth for the validation of cell segmentation methods. The cell simulation was evaluated twofold: (1) An expert observer study shows that the proposed approach generates realistic fluorescent cell micrograph simulations. (2) An automated segmentation pipeline on the simulated fluorescent cell micrographs reproduces segmentation performances of that pipeline on real fluorescent cell micrographs.ConclusionThe proposed simulation approach produces realistic fluorescent cell micrographs with corresponding ground truth. The simulated data is suited to evaluate image segmentation pipelines more efficiently and reproducibly than it is possible on manually annotated real micrographs.

Automated segmentation of brain cells for clonal analyses in fluorescence microscopy images

Segmentation is a fundamental process in microscopic cell image analysis. With the advent of recent advances in deep learning, more accurate and high-throughput cell segmentation has become feasible. However, most existing deep learning-based cell segmentation algorithms require fully annotated ground-truth cell labels, which are time-consuming and labor-intensive to generate. In this paper, we introduce Scribble2Label, a novel weakly-supervised cell segmentation framework that exploits only a handful of scribble annotations without full segmentation labels. The core idea is to combine pseudo-labeling and label filtering to generate reliable labels from weak supervision. For this, we leverage the consistency of predictions by iteratively averaging the predictions to improve pseudo labels. We demonstrate the performance of Scribble2Label by comparing it to several state-of-the-art cell segmentation methods with various cell image modalities, including bright-field, fluorescence, and electron microscopy. We also show that our method performs robustly across different levels of scribble details, which confirms that only a few scribble annotations are required in real-use cases.

BackgroundMany cell lines currently used in medical research, such as cancer cells or stem cells, grow in confluent sheets or colonies. The biology of individual cells provide valuable information, thus the separation of touching cells in these microscopy images is critical for counting, identification and measurement of individual cells. Over-segmentation of single cells continues to be a major problem for methods based on morphological watershed due to the high level of noise in microscopy cell images. There is a need for a new segmentation method that is robust over a wide variety of biological images and can accurately separate individual cells even in challenging datasets such as confluent sheets or colonies.ResultsWe present a new automated segmentation method called FogBank that accurately separates cells when confluent and touching each other. This technique is successfully applied to phase contrast, bright field, fluorescence microscopy and binary images. The method is based on morphological watershed principles with two new features to improve accuracy and minimize over-segmentation.First, FogBank uses histogram binning to quantize pixel intensities which minimizes the image noise that causes over-segmentation. Second, FogBank uses a geodesic distance mask derived from raw images to detect the shapes of individual cells, in contrast to the more linear cell edges that other watershed-like algorithms produce.We evaluated the segmentation accuracy against manually segmented datasets using two metrics. FogBank achieved segmentation accuracy on the order of 0.75 (1 being a perfect match). We compared our method with other available segmentation techniques in term of achieved performance over the reference data sets. FogBank outperformed all related algorithms. The accuracy has also been visually verified on data sets with 14 cell lines across 3 imaging modalities leading to 876 segmentation evaluation images.ConclusionsFogBank produces single cell segmentation from confluent cell sheets with high accuracy. It can be applied to microscopy images of multiple cell lines and a variety of imaging modalities. The code for the segmentation method is available as open-source and includes a Graphical User Interface for user friendly execution.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-014-0431-x) contains supplementary material, which is available to authorized users.

In this paper, we proposed a method based on the Freeman chain code to segment and count rhesus choroid-retinal vascular endothelial cells (RF/6A) automatically for fluorescence microscopy images. The proposed method consists of four main steps. First, a threshold filter and morphological transform were applied to reduce the noise. Second, the boundary information was used to generate the Freeman chain codes. Third, the concave points were found based on the relationship between the difference of the chain code and the curvature. Finally, cells segmentation and counting were completed based on the characteristics of the number of the concave points, the area and shape of the cells. The proposed method was tested on 100 fluorescence microscopic cell images, and the average true positive rate (TPR) is 98.13% and the average false positive rate (FPR) is 4.47%, respectively. The preliminary results showed the feasibility and efficiency of the proposed method.

BackgroundFilopodia are small cellular projections that help cells to move through and sense their environment. Filopodia play crucial roles in processes such as development and wound-healing. Also, increases in filopodia number or size are characteristic of many invasive cancers and are correlated with increased rates of metastasis in mouse experiments. Thus, one possible route to developing anti-metastatic therapies is to target factors that influence the filopodia system. Filopodia can be detected by eye using confocal fluorescence microscopy, and they can be manually annotated in images to quantify filopodia parameters. Although this approach is accurate, it is slow, tedious and not entirely objective. Manual detection is a significant barrier to the discovery and quantification of new factors that influence the filopodia system.ResultsHere, we present FiloDetect, an automated tool for detecting, counting and measuring the length of filopodia in fluorescence microscopy images. The method first segments the cell from the background, using a modified triangle threshold method, and then extracts the filopodia using a series of morphological operations. We verified the accuracy of FiloDetect on Rat2 and B16F1 cell images from three different labs, showing that per-cell filopodia counts and length estimates are highly correlated with the manual annotations. We then used FiloDetect to assess the role of a lipid kinase on filopodia production in breast cancer cells. Experimental results show that PI4KIII β expression leads to an increase in filopodia number and length, suggesting that PI4KIII β is involved in driving filopodia production.ConclusionFiloDetect provides accurate and objective quantification of filopodia in microscopy images, and will enable large scale comparative studies to assess the effects of different genetic and chemical perturbations on filopodia production in different cell types, including cancer cell lines.

Fluorescence microscopy imaging is an important tool in modern biological research, allowing insights into the processes of biological systems. Automated image analysis algorithms help in extracting information from these images. Validation of the automated algorithms can be done with ground truth data based on manual annotations, or using synthetic data with known ground truth. Synthetic data avoids the need to annotate manually large datasets but may lack important characteristics of the real data. In this paper, we present a framework for the generation of realistic synthetic fluorescence microscopy image sequences of cells, based on the simulation of spots with realistic motion models, noise models, and with the use of real background from microscopy images. Our framework aims to close the gap between real and synthetic image sequences. To study the effect of real backgrounds, we compared three spot detection methods using our synthetic image sequences. The results show that the real background influences spot detection, reducing the effectiveness of the spot detection algorithms, indicating the value of synthetic images with a realistic background in system validation.

Erythrocytes (RBC) are the most common type of blood cell. These cells are responsible for the delivery of oxygen to body tissues. The abnormality in erythrocyte cell affects the physical properties of red cell. It may also decrease the life span of red blood cells which may lead to stroke, anemia and other fatal diseases. Until now, Manual techniques are in practiced for diagnosis of blood cell’s diseases. However, this traditional method is tedious, time consuming and subject to sampling error. The accuracy of manual method depends on the expertise of the expert, while the accuracy of automated analyzer depends on the segmentation of objects in microscopic image of blood cell. Despite numerous efforts made for accurate blood cells image segmentation and cell counting in the literature. Still accurate segmentation is difficult due to the complexity of overlapping objects and shapes in microscopic images of blood cells. In this paper we have proposed a novel method for the segmentation of blood cells. We have used wiener filter along with Curvelet transform for image enhancement and noise removal. The snake algorithm and Gram-Schmidt orthogonalization have applied for boundary detection and image segmentation, respectively.KeywordsRBCSEMSegmentationWiener filterCurvelet

This paper proposes an accurate 3D segmentation method for visualization and quantitative analysis of differentiation activities of mouse embryonic stem (ES) cells using time-lapse confocal fluorescence microscopy images. One of critical tasks in ES cell segmentation arises due to that ES cell nuclei are often close to each other. Several segmentation methods by convexities and concavities on cell or nucleus contours to detect possible touching cells or nuclei were proposed. Comparing to image processing methods, these methods are more accurate in some conditions, however, still cannot detect touching nuclei without concavities on nucleus contours. Our method uses the nucleus size and convex, concave, strait and extrusion features on nucleus contour to touch a boundary between touching cell nuclei in 2D slices and interslices. Experimental results show our method can well detect touching boundaries of 2D and 3D nucleus for confocal microscopy images of mouse ES cells in an early stage of differentiating into neural progenitor cells. Based on the accurate ES cell segmentation, cell activities (velocities and shape changes) during differentiation can be accurately visualized and quantitatively analyzed.

The gold-standard for brain segmentation is expert manual segmentation, which is time-consuming and prone to low inter-rater reliability. Work refining fully-automated segmentation pipelines has resulted in many providing fast, reliable and accurate segmentations - though performances can vary. QyScore® is a medical imaging platform, designed for grey and white matter segmentation. We compared its performance to other state-of-the-art segmentation pipelines, being FreeSurfer, FSL, and ANTs. 54 T1-weighted images were used for manual and automatic segmentations of the whole brain grey matter (n=30), whole brain white matter (n=30), hippocampi (n=48), amygdalae (n=48), brainstem (n=49), cerebellum (n=49), caudate (n=49), putamen (n=49), thalamus (n=49), globus pallidus (n=49), and lateral ventricles (n=49) (Table 1). Automated segmentations were produced by QyScore® v1.13, FreeSurfer v7.4.1, FSL v6.0.6.2, and ANTs v2.5, using default parameters, followed by parameter optimization in the instance of preprocessing failures. Consensus manual segmentations, created by three expert neuroradiologists, were used as ground truth. Dice similarity coefficients (DSC) between consensus manual and automated segmentations were used to compare the accuracy of each segmentation pipeline. QyScore® was the best performing segmentation pipeline for whole brain grey matter, whole brain white matter, brainstem, amygdala, putamen, and lateral ventricles (Minimum-Maximum DSC QyScore®:0.79-0.93; ANTs:0.73-0.85; FreeSurfer:0.61-0.82; FSL:0.75-0.83), pbonf<0.001; Figure 1. FSL performed the best for hippocampal segmentation (DSC=0.83), followed by QyScore® (DSC=0.82), and FreeSurfer (DSC=0.67), pbonf<0.05. There were no other significant differences in segmentation accuracy. Initial automated preprocessing failed and required parameter optimization for 46.3% of subjects when using FreeSurfer, and 11.1% of subjects when using FSL. The preprocessing of one subject failed for FreeSurfer and FSL and were not recoverable without manual intervention. All subjects were preprocessed without error, using initial default parameters, by QyScore® and ANTs. QyScore® showed excellent segmentation accuracy across all brain regions, outperforming the other segmentation pipelines on the majority of regions. Alongside ANTs, it was the most robust pipeline in processing the wide variety of medical images without error. QyScore®'s high segmentation accuracy and robust pipeline demonstrate its utility in both healthcare and clinical trial settings, for supporting both clinical diagnosis and monitoring.

BackgroundThe gold‐standard for brain segmentation is expert manual segmentation, which is time‐consuming and prone to low inter‐rater reliability. Work refining fully‐automated segmentation pipelines has resulted in many providing fast, reliable and accurate segmentations ‐ though performances can vary. QyScore® is a medical imaging platform, designed for grey and white matter segmentation. We compared its performance to other state‐of‐the‐art segmentation pipelines, being FreeSurfer, FSL, and ANTs.Method54 T1‐weighted images were used for manual and automatic segmentations of the whole brain grey matter (n = 30), whole brain white matter (n = 30), hippocampi (n = 48), amygdalae (n = 48), brainstem (n = 49), cerebellum (n = 49), caudate (n = 49), putamen (n = 49), thalamus (n = 49), globus pallidus (n = 49), and lateral ventricles (n = 49) (Table 1). Automated segmentations were produced by QyScore® v1.13, FreeSurfer v7.4.1, FSL v6.0.6.2, and ANTs v2.5, using default parameters, followed by parameter optimization in the instance of preprocessing failures. Consensus manual segmentations, created by three expert neuroradiologists, were used as ground truth. Dice similarity coefficients (DSC) between consensus manual and automated segmentations were used to compare the accuracy of each segmentation pipeline.ResultQyScore® was the best performing segmentation pipeline for whole brain grey matter, whole brain white matter, brainstem, amygdala, putamen, and lateral ventricles (Minimum‐Maximum DSC QyScore®:0.79‐0.93; ANTs:0.73‐0.85; FreeSurfer:0.61‐0.82; FSL:0.75‐0.83), pbonf<0.001; Figure 1. FSL performed the best for hippocampal segmentation (DSC=0.83), followed by QyScore® (DSC=0.82), and FreeSurfer (DSC=0.67), pbonf<0.05. There were no other significant differences in segmentation accuracy.Initial automated preprocessing failed and required parameter optimization for 46.3% of subjects when using FreeSurfer, and 11.1% of subjects when using FSL. The preprocessing of one subject failed for FreeSurfer and FSL and were not recoverable without manual intervention. All subjects were preprocessed without error, using initial default parameters, by QyScore® and ANTs.ConclusionQyScore® showed excellent segmentation accuracy across all brain regions, outperforming the other segmentation pipelines on the majority of regions. Alongside ANTs, it was the most robust pipeline in processing the wide variety of medical images without error. QyScore®’s high segmentation accuracy and robust pipeline demonstrate its utility in both healthcare and clinical trial settings, for supporting both clinical diagnosis and monitoring.

Efficient globally optimal segmentation of cells in fluorescence microscopy images using level sets and convex energy functionals

Live imaging of cells and small organisms is an important step for understanding and analyzing biological functions via studying cellular dynamics. Cell segmentation and tracking in microscopy images are challenging tasks due mainly to embedded noise. We proposed to use an adaptive dictionary learning approach for filtering and reducing noise in fluorescent microscopy images. We applied our method to detect nuclei in noisy images from different types of datasets and the results demonstrated that our proposed algorithm had a satisfactory performance with an average sensitivity of 99.1%, precision of 92.4% and f-measure of 95.6%. As a result, this method is a promising preprocessing tool for detecting nuclei in noisy microscopy images.

The tracking of moving biological cells in time-lapse video sequences is fundamental to further understanding biological processes. Automatic cell tracking techniques require accurate cell image segmentation; however, current segmentation techniques are susceptible to errors due to non-ideal but realistic cell image conditions, including low contrast typical of cell microscopic images. This paper proposes a novel image pre-processing technique to enhance the low grayscale image contrast for improved cell image segmentation accuracy. A shifted bi-Gaussian model is matched to the original cell image intensity histogram for greater differentiation between the cell foreground and image background, whilst maintaining the original intensity histogram shape. Experiments conducted on a stem cell time-lapse image database show up to 33% improved segmentation accuracy, in some frames (partially or completely) detecting cells that manual ground-truth and/or existing segmentation approaches fail to identify.

Segmentation of the masseter muscle (MM) on cone-beam computed tomography (CBCT) is challenging due to the lack of sufficient soft-tissue contrast. Moreover, manual segmentation is laborious and time-consuming. The purpose of this study was to propose a deep learning-based automatic approach to accurately segment the MM from CBCT under the refinement of high-quality paired computed tomography (CT). Fifty independent CBCT and 42 clinically hard-to-obtain paired CBCT and CT were manually annotated by two observers. A 3D U-shape network was carefully designed to segment the MM effectively. Manual annotations on CT were set as the ground truth. Additionally, an extra five CT and five CBCT auto-segmentation results were revised by one oral and maxillofacial anatomy expert to evaluate their clinical suitability. CBCT auto-segmentation results were comparable to the CT counterparts and significantly improved the similarity with the ground truth compared with manual annotations on CBCT. The automatic approach was more than 332 times shorter than that of a human operation. Only 0.52% of the manual revision fraction was required. This automatic model could simultaneously and accurately segment the MM structures on CBCT and CT, which can improve clinical efficiency and efficacy, and provide critical information for personalized treatment and long-term follow-up.