Tissue Segmentation Research Articles

Abstract 5222: AI-driven, mIF-based cell-omics reveals spatially resolved cell signature for outcome prediction in NSCLC patients

Abstract Background: Despite the rapidly increasing number of new targeted and immunotherapeutic options over the past two decades, the prognosis of patients with NSCLC, even with early-stage tumors, is still poor and novel biomarkers are needed to better stratify patients in terms of survival and treatment response. A novel approach is to gain a holistic understanding of the cellular composition and formation of the tumor microenvironment (TME). Therefore, we developed a miF-based, AI-driven approach for spatially resolved TME characterization at the cellular level and used this to successfully predict clinical outcome. Methods: We assembled a large bicentric real-world sample group of 1168 patients with resected NSCLC from the Charite and the University Hospital Cologne. For tissue microarray construction, four 1.5 mm tissue cores were punched from each formalin-fixed and paraffin-embedded tumor block. Sections were stained with a 12-plex IF panel followed by H&E staining. All stains were scanned and co-registered with single cell accuracy. Next, we trained a H&E-based tissue segmentation model to detect the different tumor regions: carcinoma, stroma, and necrosis. In addition, we developed a nucleus-based cell detection model, and 12 cell classification models to categorize each detected cell by single-miF channels. Different cell phenotypes were derived from the marker-specific cell classifications. Finally, we trained a model on the Charité cohort using the spatially resolved cell readouts, spot-wise phenotype log-density, co-clustering of marker expression, and frequency of co-occurrence of marker expression through Delaunay triangulation, to predict patient survival on the Cologne cohort. Results: The tissue segmentation model achieved a macro averaged F1 score of 92%. The cell detection model identified a total of 53 million cells that were classified marker-wise with an F1 score of at least 95% on hold-out data. Our final prediction model identified a stable spatially resolved cell signature, consisting of 10 different characteristic cell neighborhood niches, which could be used to predict overall patient survival. The model trained on the Charite cohort was validated with the Cologne cohort and achieved a high performance (C-score of 71). In comparison, the UICC8 stage and the immunoscore (CD20+CD3+/carcinoma cell ratio), which were used as a baseline, achieved C-scores of 63 and 54, respectively. Conclusions: The combination of our large real-world clinical cohort, multiplex panel, and automated AI approach enabled a broad spatially resolved exploration of the TME in NSCLC at single cell resolution. Our model identified a specific cell neighborhood signature predictive of patient survival outperforming the commonly used prognostic scores, UICC8 stage and immunoscore. This allows for an improved patient stratification with potential implication for therapy selection. Citation Format: Simon Schallenberg, Gabriel Dernbach, Sharon Ruane, Cornelius Böhm, Lukas Ruff, Kai Standvoss, Sandip Ghosh, Marco Frentsch, Mihnea Dragomir, Rebecca Fritz, Ines Koch, Corinna Friedrich, Il-Kang Na, Sabine Merkelbach-Bruse, Alexander Quaas, Nikolaj Frost, Kyrill Boschung, Winfried Randerath, Georg Schlachtenberger, Matthias Heldwein, Ulrich Keilholz, Khosro Hekmat, Jens Rückert, Reinhard Büttner, Angela Vasaturo, David Horst, Maximilian Alber, Frederick Klauschen. AI-driven, mIF-based cell-omics reveals spatially resolved cell signature for outcome prediction in NSCLC patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5222.

Abstract 6211: Clinical and prognostic assessment of the lncRNA ZFAS1-related gene signature in colorectal cancer

Abstract Colorectal cancer (CRC) is one of the leading causes of cancer death with a rapidly rising incidence rate in younger adults. It is projected to be the leading cause of death in adults under 50 years old by 2030. Although progress over the last couple of decades has improved the clinical management of patients affected by CRC, there is a lack of accurate biomarkers to stratify CRC patients. Among several other biomolecules, lncRNAs (long non-coding RNAs) are non-coding transcripts under 200 nucleotides that can regulate gene expression. Furthermore, lncRNAs are important players in oncogenesis and cancer progression and have the potential to be used as reliable biomarkers. Through this study, we explored the clinical relevance of lncRNAs in colorectal cancer. To analyze the distribution of lncRNAs, we utilized de-identified FFPE (formalin-fixed paraffin-embedded) blocks from Augusta University (n = 88). FFPE blocks were sectioned to extract tissue segments and were subsequently stained using H&E (hematoxylin and eosin dyes) to identify the tumor and adjacent normal regions. The quantification of the gene expression was performed using a medium-throughput NanoString platform. Further analyses were performed using clinicopathological features of the institutional cohort and an independent Cancer Genome Atlas (TCGA) dataset. TCGA contains genomic data along with clinical and molecular characterization of over 11,000 samples of various lineages. In the institutional cohort, ZFAS1 showed higher gene expression in the low survival group for stage II and III colorectal cancer patients, while no significant difference was observed in stage I and IV patients. ZFAS1 expression did not significantly correlate with other clinicopathological parameters, including race, alcohol consumption, tobacco consumption, age, grade, and family history of cancer. In an orthogonal TCGA analysis, higher ZFAS1 expression was associated with genomic perturbations, including significantly altered tumor mutational burden (TMB), aneuploidy, fractional genome alteration, and microsatellite instability. Further, a 21-gene signature was developed for prognostication using gene co-expression network analysis. The gene signature was significantly associated with overall and disease-free survival in the TCGA-CRC cohort. Immune deconvolution analysis of the RNA sequencing data revealed that higher expression of the 21-gene signature was associated with increased infiltration of regulatory T cells while being inversely correlated with the overall level of T cells. In conclusion, our study identified the clinicopathological role of ZFAS1 lncRNA and its interaction genes with prognostic significance in colorectal cancer. Further investigation into the therapeutic relevance of the ZFAS1-related gene signature may prove invaluable in improving the clinical outcomes of colorectal cancer patients. Citation Format: Pankaj Kumar Ahluwalia, Tiffanie Leeman, Ashis Mondal, Ashutosh Vashisht, Harmanpreet Singh, Nivin Omar, Kimya Jones, Ravindra Kolhe. Clinical and prognostic assessment of the lncRNA ZFAS1-related gene signature in colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6211.

Quantitative Proteomic Analysis Identifying and Evaluating TRAF6 and IL-8 as Potential Diagnostic Biomarkers in Neonatal Patients with Necrotizing Enterocolitis.

Necrotizing enterocolitis (NEC) is a common gastrointestinal complication in premature infants, resulting in high morbidity and mortality, and its early detection is crucial for accurate treatment and outcome prediction. Extensive research has demonstrated a clear correlation between NEC and extremely low birth weight, degree of preterm, formula feeding, infection, hypoxic/ischemic damage, and intestinal dysbiosis. The development of noninvasive biomarkers of NEC from stool, urine, and serum has attracted a great deal of interest because to these clinical connections and the quest for a deeper knowledge of disease pathophysiology. Therefore, this study aims to identify protein expression patterns in NEC and discover innovative diagnostic biomarkers. In this study, we recruited five patients diagnosed with NEC and paired necrotic segments of intestinal tissue with adjacent normal segments of intestine to form experimental and control groups. Quantitative proteomics tandem mass tagging (TMT) labeling technique was used to detect and quantify the proteins, and the expression levels of the candidate biomarkers in the intestinal tissues were further determined by quantitative polymerase chain reaction (RT-qPCR), Western blot analysis, Immunofluorescence methods and enzyme-linked immunosorbent assay (ELISA). A total of 6880 proteins were identified and quantified in patients with NEC. A significant disparity in protein expression was observed between necrotic and normal segments of intestinal tissue in NEC patients. A total of 55 proteins were found to be upregulated, and 40 proteins were found to be downregulated in NEC patients when using a p-value of < 0.05, and an absolute fold change of > 1.2 for analysis. GO function enrichment analysis showed the positive regulation of significant biological processes such as mitochondrial organization, vasoconstriction, rRNA catabolism, fluid shear stress response, and glycerol ether biosynthesis processes. Enrichment analysis also revealed essential functions such as ligand-gated ion channel activity, potassium channel activity, ligand-gated cation channel activity, ligand-gated ion channel activity, and ligand-gated channel activity, including molecular functions such as ligand-gated ion channel activity and mitotic events in this comparative group. Significant changes were found in endomembrane protein complex, membrane fraction, mitochondrial membrane fraction, membrane components, membrane intrinsic components, and other localized proteins. Additional validation of intestinal tissue and serum revealed a substantial increase in TRAF6 (tumor necrosis factor receptor-associated factor 6) and IL-8(Interleukin-8, CXCL8). The quantitative proteomic TMT method can effectively detect proteins with differential expression in the intestinal tissues of NEC patients. Proteins TRAF6 and CXCL8/IL-8 are significantly upregulated in the intestinal tissues and serum samples of patients and may serve as valuable predictor factors for NEC's early diagnosis.

Effective classification of alzheimer disease based on image tractography framework utilizing GM-ABC-NN

A progressive neurological disorder that leads to memory loss as well as cognitive decline is called Alzheimer’s Disease (AD). Grounded on various factors like symptoms, biomarkers, or disease stages, the classification of AD is done. Yet, the prevailing techniques pose challenges in capturing accurate AD, which results in misdiagnosis or delayed treatment. To overcome these issues, an effective framework for the classification of AD based on image tractography using several algorithms is proposed in this article. Primarily, the dataset comprising AD images undergoes image transformation. Then, to enhance the MRI image quality, the transformed images are pre-processed using Block Matching 3D Filtering (BM3D) and Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithms. Afterward, image tractography followed by brain tissue segmentation is done utilizing the Cauchy-Steiner Weighted Fuzzy C Means (CauSt-FCM) algorithm. Then, by utilizing the Cosine Non-Negative Kernel Regression (CosNNK) algorithm, graph construction is done. Lastly, for the feature selection and validation process, the Premature Singer Archerfish Hunting Optimizer (PreS-AHO) algorithm and Analysis of Variance (ANOVA) analysis are used. Thereafter, the AD stages are classified by using the Gaussian Mixture Alpha-Beta Convolutional Neural Network (GM-ABC-NN) algorithm, The proposed system attains superior accuracy (97.56%), precision (97.55%), and recall (97.57%).

A polar transformation augmentation approach for enhancing mammary gland segmentation in ultrasound images

Environmental factors can detrimentally affect mammary gland development, leading to negative impacts on milk secretion in mammals. Ultrasonography serves as a non-invasive and non-destructive method for assessing mammary gland characteristics and development. Deep learning approaches enable automated monitoring of mammary gland development, though they typically require large, labeled datasets that may be limited by data collection constraints. This study aimed to develop and evaluate a polar transformation-based augmentation strategy to enhance the performance of deep learning algorithms for mammary gland segmentation in small datasets. We collected 405 ultrasound images of mammary glands (front and rear quarters) from 29 crossbred F1 Holstein x Gyr calves aged 5 to 11 weeks. The parenchyma tissue in these images was manually annotated using the VGG Image Annotator. A leave-one-animal-out cross-validation approach was employed to train the semantic segmentation algorithm. In this approach, all images from one calf were used as a testing set, and images from the remaining 28 calves were used for training in each of the 29 iterations. Our proposed method involved utilizing a polar transform technique for data augmentation in ultrasound images and the PSPNet deep learning algorithm for image segmentation. The average F1-score on the testing set was 54% in week 1, 70% in week 2, and 75% in week 3. Our findings revealed that the algorithm’s performance was suboptimal for images with very small parenchyma (week 1). However, as the mammary gland developed, the identification and segmentation of parenchymal tissue significantly improved. The performance of deep learning algorithms in segmenting small tissues could potentially be enhanced by using larger datasets and higher resolution images. In conclusion, our study demonstrates that polar transformation is an effective strategy for augmenting mammary gland ultrasound images, which in turn improves the performance of deep neural networks in segmenting parenchymal tissue.

Rapid and Label-Free Histopathology of Oral Lesions Using Deep Learning Applied to Optical and Infrared Spectroscopic Imaging Data.

Oral potentially malignant disorders (OPMDs) are precursors to over 80% of oral cancers. Hematoxylin and eosin (H&E) staining, followed by pathologist interpretation of tissue and cellular morphology, is the current gold standard for diagnosis. However, this method is qualitative, can result in errors during the multi-step diagnostic process, and results may have significant inter-observer variability. Chemical imaging (CI) offers a promising alternative, wherein label-free imaging is used to record both the morphology and the composition of tissue and artificial intelligence (AI) is used to objectively assign histologic information. Here, we employ quantum cascade laser (QCL)-based discrete frequency infrared (DFIR) chemical imaging to record data from oral tissues. In this proof-of-concept study, we focused on achieving tissue segmentation into three classes (connective tissue, dysplastic epithelium, and normal epithelium) using a convolutional neural network (CNN) applied to three bands of label-free DFIR data with paired darkfield visible imaging. Using pathologist-annotated H&E images as the ground truth, we demonstrate results that are 94.5% accurate with the ground truth using combined information from IR and darkfield microscopy in a deep learning framework. This chemical-imaging-based workflow for OPMD classification has the potential to enhance the efficiency and accuracy of clinical oral precancer diagnosis.

Thoracic aortic microcalcification activity in combined positron emission tomography and magnetic resonance imaging

IntroductionNon-invasive detection of pathological changes in thoracic aortic disease remains an unmet clinical need particularly for patients with congenital heart disease. Positron emission tomography combined with magnetic resonance imaging (PET-MRI) could provide a valuable low-radiation method of aortic surveillance in high-risk groups. Quantification of aortic microcalcification activity using sodium [18F]fluoride holds promise in the assessment of thoracic aortopathies. We sought to evaluate aortic sodium [18F]fluoride uptake in PET-MRI using three methods of attenuation correction compared to positron emission tomography computed tomography (PET-CT) in patients with bicuspid aortic valve,MethodsThirty asymptomatic patients under surveillance for bicuspid aortic valve disease underwent sodium [18F]fluoride PET-CT and PET-MRI of the ascending thoracic aorta during a single visit. PET-MRI data were reconstructed using three iterations of attenuation correction (Dixon, radial gradient recalled echo with two [RadialVIBE-2] or four [RadialVIBE-4] tissue segmentation). Images were qualitatively and quantitatively analysed for aortic sodium [18F]fluoride uptake on PET-CT and PET-MRI.ResultsAortic sodium [18F]fluoride uptake on PET-MRI was visually comparable with PET-CT using each reconstruction and total aortic standardised uptake values on PET-CT strongly correlated with each PET-MRI attenuation correction method (Dixon R = 0.70; RadialVIBE-2 R = 0.63; RadialVIBE-4 R = 0.64; p < 0.001 for all). Breathing related artefact between soft tissue and lung were detected using Dixon and RadialVIBE-4 but not RadialVIBE-2 reconstructions, with the presence of this artefact adjacent to the atria leading to variations in blood pool activity estimates. Consequently, quantitative agreements between radiotracer activity on PET-CT and PET-MRI were most consistent with RadialVIBE-2.ConclusionAscending aortic microcalcification analysis in PET-MRI is feasible with comparable findings to PET-CT. RadialVIBE-2 tissue attenuation correction correlates best with the reference standard of PET-CT and is less susceptible to artefact. There remain challenges in segmenting tissue types in PET-MRI reconstructions, and improved attenuation correction methods are required.

ResDAC-Net: a novel pancreas segmentation model utilizing residual double asymmetric spatial kernels

The pancreas not only is situated in a complex abdominal background but is also surrounded by other abdominal organs and adipose tissue, resulting in blurred organ boundaries. Accurate segmentation of pancreatic tissue is crucial for computer-aided diagnosis systems, as it can be used for surgical planning, navigation, and assessment of organs. In the light of this, the current paper proposes a novel Residual Double Asymmetric Convolution Network (ResDAC-Net) model. Firstly, newly designed ResDAC blocks are used to highlight pancreatic features. Secondly, the feature fusion between adjacent encoding layers fully utilizes the low-level and deep-level features extracted by the ResDAC blocks. Finally, parallel dilated convolutions are employed to increase the receptive field to capture multiscale spatial information. ResDAC-Net is highly compatible to the existing state-of-the-art models, according to three (out of four) evaluation metrics, including the two main ones used for segmentation performance evaluation (i.e., DSC and Jaccard index).Graphical abstract

A publicly available deep learning model and dataset for segmentation of breast, fibroglandular tissue, and vessels in breast MRI

Breast density, or the amount of fibroglandular tissue (FGT) relative to the overall breast volume, increases the risk of developing breast cancer. Although previous studies have utilized deep learning to assess breast density, the limited public availability of data and quantitative tools hinders the development of better assessment tools. Our objective was to (1) create and share a large dataset of pixel-wise annotations according to well-defined criteria, and (2) develop, evaluate, and share an automated segmentation method for breast, FGT, and blood vessels using convolutional neural networks. We used the Duke Breast Cancer MRI dataset to randomly select 100 MRI studies and manually annotated the breast, FGT, and blood vessels for each study. Model performance was evaluated using the Dice similarity coefficient (DSC). The model achieved DSC values of 0.92 for breast, 0.86 for FGT, and 0.65 for blood vessels on the test set. The correlation between our model’s predicted breast density and the manually generated masks was 0.95. The correlation between the predicted breast density and qualitative radiologist assessment was 0.75. Our automated models can accurately segment breast, FGT, and blood vessels using pre-contrast breast MRI data. The data and the models were made publicly available.

Pan-cancer image segmentation based on feature pyramids and Mask R-CNN framework.

Cancer, a disease with a high mortality rate, poses a great threat to patients' physical and mental health and can lead to huge medical costs and emotional damage. With the continuous development of artificial intelligence technologies, deep learning-based cancer image segmentation techniques are becoming increasingly important in cancer detection and accurate diagnosis. However, in segmentation tasks, there are differences in efficiency between large and small objects and limited segmentation effects on objects of individual sizes. The previous segmentation frameworks still have room for improvement in multi-scale collaboration when segmenting objects. This paper proposes a method to train a deep learning segmentation framework using a feature pyramid processing dataset to improve the average precision (AP) index, and realizes multi-scale cooperation in target segmentation. Pan-Cancer Histology Dataset for Nuclei Instance Segmentation and Classification (PanNuke) dataset was selected to include approximately 7500 pathology images with cells from 19 different types of tissues, including five classifications of cancer, non-cancer, inflammation, death, and connective tissue. First, the method uses whole-slide images in the pan-cancer histology dataset for nuclei instance segmentation and classification (PanNuke) dataset, combined with the mask region convolutional neural network (Mask R-CNN) segmentation framework and improved loss function to segment and detect each cellular tissue in cancerous sections. Second, to address the problem of non-synergistic object segmentation at different scales in cancerous tissue segmentation, a scheme using feature pyramids to process the dataset was adopted as part of the feature extraction module. Extensive experimental results on this dataset show that the method in this paper yields 0.269 AP and a boost of about 4% compared to the original Mask R-CNN framework. It is effective and feasible to use feature pyramid to process data set to improve the effect of medical image segmentation.

Community Structure and Diversity of Endophytic Fungi in Cultivated Polygala crotalarioides at Two Different Growth Stages Based on Culture-Independent and Culture-Based Methods.

Polygala crotalarioides, a perennial herbaceous plant found in southwest China, has the potential to be used in the treatment of Alzheimer's disease. Endophytic fungi that reside within medicinal herbs play an important ecological role in their host plants and can serve as a valuable source for identifying active components. However, little is known about the diversity, and structure of endophytic fungi in P. crotalarioides. In this study, we investigated the community structure and diversity of endophytic fungi in the leaves, stems, and roots of P. crotalarioides at both 1- and 2-year-growth stages using a modern culture-independent method using both culture-independent (high-throughput sequencing, HTS) and culture-based methods. Using HTS, our results revealed that the richness and diversity of endophytic fungi in P. crotalarioides varied depending on the organs and growth stages. Specifically, stems and leaves exhibited significantly higher diversity compared to roots. Additionally, the highest diversity of endophytic fungi was observed in the stems of the 2-year-old plants. At the genus level, Fusarium, Colletotrichum, and Phoma were the most abundant endophytic fungi in 1-year-old samples, while Cercospora, Apiotrichum, and Fusarium were prevalent in 2-year-old samples. A total of 55 endophytic fungal strains belonging to two phyla and 24 genera were isolated from 150 plant tissue segments using culture-based methods. The anti-acetylcholinesterase activity of these isolates was evaluated in vitro and five of them, Phialophora mustea PCAM010, Diaporthe nobilis PCBM027, Fusarium oxysporum LP41, F. oxysporum SR60, and Phoma herbarum SM81, showed strong activity (>50% inhibition rate). These findings will serve as a theoretical basis and practical guide for comprehending the structural composition, biological diversity and bioactivity of endophytic fungi in P. crotalarioides.

PETS-Nets: Joint Pose Estimation and Tissue Segmentation of Fetal Brains Using Anatomy-Guided Networks.

Fetal Magnetic Resonance Imaging (MRI) is challenged by fetal movements and maternal breathing. Although fast MRI sequences allow artifact free acquisition of individual 2D slices, motion frequently occurs in the acquisition of spatially adjacent slices. Motion correction for each slice is thus critical for the reconstruction of 3D fetal brain MRI. In this paper, we propose a novel multi-task learning framework that adopts a coarse-to-fine strategy to jointly learn the pose estimation parameters for motion correction and tissue segmentation map of each slice in fetal MRI. Particularly, we design a regression-based segmentation loss as a deep supervision to learn anatomically more meaningful features for pose estimation and segmentation. In the coarse stage, a U-Net-like network learns the features shared for both tasks. In the refinement stage, to fully utilize the anatomical information, signed distance maps constructed from the coarse segmentation are introduced to guide the feature learning for both tasks. Finally, iterative incorporation of the signed distance maps further improves the performance of both regression and segmentation progressively. Experimental results of cross-validation across two different fetal datasets acquired with different scanners and imaging protocols demonstrate the effectiveness of the proposed method in reducing the pose estimation error and obtaining superior tissue segmentation results simultaneously, compared with state-of-the-art methods.

Effect of topical steroid on soft tissue swelling following anterior cervical discectomy and fusion.

Anterior cervical discectomy and fusion is the most commonly used surgical approach for treating cervical spine conditions, but it can often lead to postoperative swallowing difficulties. To retrospectively assess the effects of topical triamcinolone acetonide in the anterior cervical surgery on swallowing function. In this study, a retrospective design was used to select patients aged 18 years and older who were diagnosed with cervical spondylosis and required anterior cervical discectomy and fusion. Among them, the patients in the experimental group used triamcinolone acetonide topically in front of the plate during surgery, and the control group was the patients who did not use triamcinolone acetonide. The sex, age, operation time, operation segment, and preoperative soft tissue area were compared between the two groups. There were no significant differences in gender, age, operation time, and segment between the two groups. For the preoperative soft tissue area, triamcinolone acetonide was significantly lower than in the control group (P < 0.05). The retrospective results of this study support that topical triamcinolone acetonide as a treatment in anterior cervical surgery can significantly reduce soft tissue swelling, and no effect was found on the operation time, postoperative blood loss, and segment. These findings provide an important basis for clinical care teams to make treatment decisions and confirm the effectiveness of triamcinolone acetonide in improving swallowing function. However, there was a possibility of information collection and selection bias due to the limitations of retrospective studies. To confirm and further advance the use of this treatment, more rigorous prospective randomized controlled trials are recommended to validate these preliminary results.

310 (PB-126) Poster - Breast tissue segmentation in MR images using deep-learning

310 (PB-126) Poster - Breast tissue segmentation in MR images using deep-learning

An analysis on the effect of body tissues and surgical tools on workflow recognition in first person surgical videos.

Analysis of operative fields is expected to aid in estimating procedural workflow and evaluating surgeons' procedural skills by considering the temporal transitions during the progression of the surgery. This study aims to propose an automatic recognition system for the procedural workflow by employing machine learning techniques to identify and distinguish elements in the operative field, including body tissues such as fat, muscle, and dermis, along with surgical tools. We conducted annotations on approximately 908 first-person-view images of breast surgery to facilitate segmentation. The annotated images were used to train a pixel-level classifier based on Mask R-CNN. To assess the impact on procedural workflow recognition, we annotated an additional 43,007 images. The network, structured on the Transformer architecture, was then trained with surgical images incorporating masks for body tissues and surgical tools. The instance segmentation of each body tissue in the segmentation phase provided insights into the trend of area transitions for each tissue. Simultaneously, the spatial features of the surgical tools were effectively captured. In regard to the accuracy of procedural workflow recognition, accounting for body tissues led to an average improvement of 3 % over the baseline. Furthermore, the inclusion of surgical tools yielded an additional increase in accuracy by 4 % compared to the baseline. In this study, we revealed the contribution of the temporal transition of the body tissues and surgical tools spatial features to recognize procedural workflow in first-person-view surgical videos. Body tissues, especially in open surgery, can be a crucial element. This study suggests that further improvements can be achieved by accurately identifying surgical tools specific to each procedural workflow step.

Application of clustering strategy for automatic segmentation of tissue regions in mass spectrometry imaging.

Mass spectrometry imaging (MSI) has been widely used in biomedical research fields. Each pixel in MSI consists of a mass spectrum that reflects the molecule feature of the tissue spot. Because MSI contains high-dimensional datasets, it is highly desired to develop computational methods for data mining and constructing tissue segmentation maps. To visualize different tissue regions based on mass spectrum features and improve the efficiency in processing enormous data, we proposed a computational strategy that consists of four procedures including preprocessing, data reduction, clustering, and quantitative validation. In this study, we examined the combination of t-distributed stochastic neighbor embedding (t-SNE) and hierarchical clustering (HC) for MSI data analysis. Using publicly available MSI datasets, one dataset of mouse urinary bladder, and one dataset of human colorectal cancer, we demonstrated that the generated tissue segmentation maps from this combination were superior to other data reduction and clustering algorithms. Using the staining image as a reference, we assessed the performance of clustering algorithms with external and internal clustering validation measures, including purity, adjusted Rand index (ARI), Davies-Bouldin index (DBI), and spatial aggregation index (SAI). The result indicated that SAI delivered excellent performance for automatic segmentation of tissue regions in MSI. We used a clustering algorithm to construct tissue automatic segmentation in MSI datasets. The performance was evaluated by comparing it with the stained image and calculating clustering validation indexes. The results indicated that SAI is important for automatic tissue segmentation in MSI, different from traditional clustering validation measures. Compared to the reports that used internal clustering validation measures such as DBI, our method offers more effective evaluation of clustering results for MSI segmentation. We envision that the proposed automatic image segmentation strategy can facilitate deep learning in molecular feature extraction and biomarker discovery for the biomedical applications of MSI.

Shape-constrained deformable brain segmentation: Methods and quantitative validation

MRI-guided neuro interventions require rapid, accurate, and reproducible segmentation of anatomical brain structures for identification of targets during surgical procedures and post-surgical evaluation of intervention efficiency. Segmentation algorithms must be validated and cleared for clinical use. This work introduces a methodology for shape-constrained deformable brain segmentation, describes the quantitative validation used for its clinical clearance, and presents a comparison with manual expert segmentation and FreeSurfer, an open source software for neuroimaging data analysis. ClearPoint Maestro is software for fully-automatic brain segmentation from T1-weighted MRI that combines a shape-constrained deformable brain model with voxel-wise tissue segmentation within the cerebral hemispheres and the cerebellum. The performance of the segmentation was validated in terms of accuracy and reproducibility. Segmentation accuracy was evaluated with respect to training data and independently traced ground truth. Segmentation reproducibility was quantified and compared with manual expert segmentation and FreeSurfer. Quantitative reproducibility analysis indicates superior performance compared to both manual expert segmentation and FreeSurfer. The shape-constrained methodology results in accurate and highly reproducible segmentation. Inherent point based-correspondence provides consistent target identification ideal for MRI-guided neuro interventions.

A color-based tumor segmentation method for clinical ex vivo breast tissue assessment utilizing a multi-contrast brightfield imaging strategy.

We demonstrate an automated two-step tumor segmentation method leveraging color information from brightfield images of fresh core needle biopsies of breast tissue. Three different color spaces (HSV, CIELAB, YCbCr) were explored for the segmentation task. By leveraging white-light and green-light images, we identified two different types of color transformations that could separate adipose from benign and tumor or cancerous tissue. We leveraged these two distinct color transformation methods in a two-step process where adipose tissue segmentation was followed by benign tissue segmentation thereby isolating the malignant region of the biopsy. Our tumor segmentation algorithm and imaging probe could highlight suspicious regions on unprocessed biopsy tissue to guide selection of areas most similar to malignant tissues for tissue pathology whether it be formalin fixed or frozen sections, expedite tissue selection for molecular testing, detect positive tumor margins, or serve an alternative to tissue pathology, in countries where these services are lacking.

Precise and Rapid Whole-Head Segmentation from Magnetic Resonance Images of Older Adults using Deep Learning.

Whole-head segmentation from Magnetic Resonance Images (MRI) establishes the foundation for individualized computational models using finite element method (FEM). This foundation paves the path for computer-aided solutions in fields, particularly in non-invasive brain stimulation. Most current automatic head segmentation tools are developed using healthy young adults. Thus, they may neglect the older population that is more prone to age-related structural decline such as brain atrophy. In this work, we present a new deep learning method called GRACE, which stands for General, Rapid, And Comprehensive whole-hEad tissue segmentation. GRACE is trained and validated on a novel dataset that consists of 177 manually corrected MR-derived reference segmentations that have undergone meticulous manual review. Each T1-weighted MRI volume is segmented into 11 tissue types, including white matter, grey matter, eyes, cerebrospinal fluid, air, blood vessel, cancellous bone, cortical bone, skin, fat, and muscle. To the best of our knowledge, this work contains the largest manually corrected dataset to date in terms of number of MRIs and segmented tissues. GRACE outperforms five freely available software tools and a traditional 3D U-Net on a five-tissue segmentation task. On this task, GRACE achieves an average Hausdorff Distance of 0.21, which exceeds the runner-up at an average Hausdorff Distance of 0.36. GRACE can segment a whole-head MRI in about 3 seconds, while the fastest software tool takes about 3 minutes. In summary, GRACE segments a spectrum of tissue types from older adults T1-MRI scans at favorable accuracy and speed. The trained GRACE model is optimized on older adult heads to enable high-precision modeling in age-related brain disorders. To support open science, the GRACE code and trained weights are made available online and open to the research community at https://github.com/lab-smile/GRACE.

SeqCorr-EUNet: A sequence correction dual-flow network for segmentation and quantification of anterior segment OCT image

The accurate segmentation of AS-OCT images is a prerequisite for the morphological details analysis of anterior segment structure and the extraction of clinical biological parameters, which play an essential role in the diagnosis, evaluation, and preoperative prognosis management of many ophthalmic diseases. Manually marking the boundaries of the anterior segment tissue is time-consuming and error-prone, with inherent speckle noise, various artifacts, and some low-quality scanned images further increasing the difficulty of the segmentation task. In this work, we propose a novel model called SeqCorr-EUNet with a dual-flow architecture based on convolutional gated recursive sequence correction for semantic segmentation and quantification of AS-OCT images. An EfficientNet encoder is employed to enhance the intra-slice features extraction ability of semantic segmentation flow. The sequence correction flow based on ConvGRU is introduced to extract inter-slice features from consecutive adjacent slices. Spatio-temporal information is fused to correct the morphological details of pre-segmentation results. And the channel attention gate is inserted into the skip-connection between encoder and decoder to enrich the contextual information and suppress the noise of irrelevant regions. Based on the segmentation results of the anterior segment structures, we achieved automatic extraction of essential clinical parameters, 3D reconstruction of the anterior chamber structure, and measurement of anterior chamber volume. The proposed SeqCorr-EUNet has been evaluated on the public AS-OCT dataset. The experimental results show that our method is competitive compared with the existing methods and significantly improves the segmentation and quantification performance of low-quality imaging structures in AS-OCT images.