Nodal Segments Research Articles

Artificial intelligence aided diagnosis of pulmonary nodules segmentation and feature extraction

To develop a high-accuracy low-dose computed tomography (LDCT) lung nodule diagnosis system by combining artificial intelligence (AI) technology with the Lung CT Screening Reporting and Data System (Lung-RADS), which can be used in the future AI-aided diagnosis of pulmonary nodules. The study comprised the following steps: (1) the best deep-learning segmentation method for pulmonary nodules was compared and selected objectively; (2) the Image Biomarker Standardization Initiative (IBSI) was used for feature extraction and to determine the best feature reduction method; and (3) a principal component analysis (PCA) and three machine learning methods were used to analyse the extracted features, and the best method was determined. The Lung Nodule Analysis 16 dataset was applied to train and test the established system in this study. The competition performance metric (CPM) score of the nodule segmentation reached 0.83, the accuracy of nodule classification was 92%, the kappa coefficient with the ground truth was 0.68, and the overall diagnostic accuracy (calculated by the nodules) was 0.75. This paper summarises a more efficient AI-assisted diagnosis process of pulmonary nodules, and has better performance compared with the previous literature. In addition, this method will be validated in a future external clinical study.

New evidence for in vitro conservation of nodal segments of the passion fruit ‘UENF Rio Dourado’ (Passiflora edulis Sims)

The germplasm of Passiflora L. is conserved through seed banks and field collections. Dormant seeds, seeds with low viability, and high-cost field collections make it difficult to maintain germplasm banks, and in vitro conservation can be a complementary alternative. The aim of this study was to investigate the survival of nodal segments of Passiflora edulis Sims ‘UENF Rio Dourado’ over 180 days of slow growth in vitro by reducing the mineral salt and sucrose concentrations and changing the incubation conditions (temperature and light intensity). The experiment was conducted in a completely randomized design with a 2 × 3 × 3 factorial arrangement consisting of two temperatures (20 ± 2°C and 27 ± 2°C), three concentrations of MSM mineral salts (100% MSM, 50% MSM, and 25% MSM), and three sucrose concentrations (10, 20, and 30 g L-1). Evaluations were performed at 60, 90, 120, 150, and 180 days for survival, number of leaves, and plant color (using a color scale where 1 = dark green, 2 = light green, and 3 = yellow). After 180 days of culture, mineral salt, sucrose concentration, and incubation temperature affected plant survival, regeneration, and acclimatization. During the 180 days, it was possible to slow the growth of nodal segments of passion fruit ‘UENF Rio Dourado’ in culture medium with 25% MSM mineral salts plus 10 g L-1 sucrose, at an average temperature of 20°C without compromising plant survival, regeneration, and acclimatization.

Development of basil ‘Dark Opal’ from explant type and plant regulators

Seedling abnormalities are a common problem in micropropagation and require investigation. This study aimed to assess the growth characteristics and morphoanatomy of stem apices and nodal segments of basil (Ocimum basilicum) ‘Dark Opal’ inoculated in medium containing combinations of 6-benzylaminopurine (BAP) and naphthaleneacetic acid (NAA). Presence of calluses, adventitious roots, and abnormal seedlings was quantified at 20 and 40 days after inoculation. Growth, anatomical, and morphological parameters were evaluated after 80 days. Explant type and regulators influenced the growth and morphoanatomy of basil ‘Dark Opal’. Treatment of stem apices with 0.05 and 0.1 mg L-1 of NAA and BAP increased leaf number, fresh weight, and dry weight and decreased the incidence of calluses and abnormalities. Hyperhydric seedlings showed morphoanatomical alterations, such as disorganization and expansion of intercellular spaces with a consequent increase in parenchymal length. Thus, the results of this research allowed optimizing and understanding the important effects of growth regulators on the development of this species, which will help in in vitro mass culture programs and in biofactories for the production of secondary metabolites.

Lung cancer classification and identification framework with automatic nodule segmentation screening using machine learning

Lung cancer classification and identification framework with automatic nodule segmentation screening using machine learning

Effect of Explant Source on Phenotypic Changes of In Vitro Grown Cannabis Plantlets over Multiple Subcultures

Drug-type cannabis is often multiplied using micropropagation methods to produce genetically uniform and disease/insect-free crops. However, micropropagated plantlets often exhibit phenotypic variation, leading to culture decline over time. In cannabis, the source of these changes remains unknown, though several factors (e.g., explant's sources and prolonged in vitro culture) can result in such phenotypical variations. The study presented herein evaluates the effects of explant sources (i.e., nodal segments derived from the basal, near-basal, middle, and apical parts of the greenhouse-grown mother plant) over multiple subcultures (4 subcultures during 235 days) on multiplication parameters and leaf morphological traits of in vitro cannabis plantlets. While initial in vitro responses were similar among explants sourced from different regions of the plant, there were significant differences in performance over the course of multiple subcultures. Specifically, explant source and/or the number of subcultures significantly impacted plantlet height, number of nodes, and canopy surface area. The explants derived from the basal and near-basal parts of the plant resulted in the tallest shoots with the greatest number of nodes, while the explants derived from the middle and apical regions led to shorter shoots with fewer nodes. Moreover, the basal-derived explants produced cannabis plantlets with shorter but wider leaves which demonstrated the potential of such explants for in vitro rejuvenation practices with minimal culture decline. This study provides new evidence into the long-term impacts of explant source in cannabis micropropagation.

Lung Nodule CT Image Segmentation Model Based on Multiscale Dense Residual Neural Network

To solve the problem of the low segmentation accuracy of lung nodule CT images using U-Net, an improved method for segmentation of lung nodules by U-Net was proposed. Initially, the dense network connection and sawtooth expanded convolution design was added to the feature extraction part, and a local residual design was adopted in the upsampling process. Finally, the effectiveness of the proposed algorithm was evaluated using the LIDC-IDRI lung nodule public dataset. The results showed that the improved algorithm had 7.03%, 14.05%, and 10.43% higher performance than the U-Net segmentation algorithm under the three loss functions of DC, MIOU, and SE, and the accuracy was 2.45% higher compared with that of U-Net. Thus, the proposed method had an effective network structure.

Nodal Segment Size and Encapsulation Matrix Affect Sugarcane Bud Emergence from Synthetic Seeds

Nodal Segment Size and Encapsulation Matrix Affect Sugarcane Bud Emergence from Synthetic Seeds

Symmetrical awareness network for cross-site ultrasound thyroid nodule segmentation.

Recent years have seen remarkable progress of learning-based methods on Ultrasound Thyroid Nodules segmentation. However, with very limited annotations, the multi-site training data from different domains makes the task remain challenging. Due to domain shift, the existing methods cannot be well generalized to the out-of-set data, which limits the practical application of deep learning in the field of medical imaging. In this work, we propose an effective domain adaptation framework which consists of a bidirectional image translation module and two symmetrical image segmentation modules. The framework improves the generalization ability of deep neural networks in medical image segmentation. The image translation module conducts the mutual conversion between the source domain and the target domain, while the symmetrical image segmentation modules perform image segmentation tasks in both domains. Besides, we utilize adversarial constraint to further bridge the domain gap in feature space. Meanwhile, a consistency loss is also utilized to make the training process more stable and efficient. Experiments on a multi-site ultrasound thyroid nodule dataset achieve 96.22% for PA and 87.06% for DSC in average, demonstrating that our method performs competitively in cross-domain generalization ability with state-of-the-art segmentation methods.

Somatic Embryogenesis of Myrciaria dubia (Kunth.) McVaugh

Myrciaria dubia (caçari or camu-camu) is a species with great relevance because of its high levels of bioactive compounds and antioxidant activity. The species is propagated mainly by seed, which leads to high genetic diversity. Producing plants that maintain their characteristics on a large scale at a low cost is a challenge for fruit species, especially in the Myrtaceae family. Therefore, this study aimed at assessing the effects of different plant growth regulators at various concentrations for the induction of somatic embryogenesis from caçari nodal segments and leaf disk explants. Two independent experiments were performed using nodal segments and leaf disks from plants grown in the greenhouse. In the first experiment, the combined effect of auxin and cytokinin at different concentrations was evaluated: 2,4-dichlorophenoxyacetic acid at 0, 1, 2, and 4 mg⋅L–1; and benzylaminopurine at 0, 0.25, 0.5, and 1 mg⋅L–1). In the second experiment, the application of different plant growth regulators (benzylaminopurine, kinetin, thidiazuron, and isopentenyl adenine) and their concentrations (0, 2, 4, and 6 mg⋅L–1) were evaluated. In both experiments, the basic culture medium was woody plant medium. Callus formation via nodal segments and leaf disks occurred in the first 30 d from cultivation and proved to be responsive to induction by 2,4-dichlorophenoxyacetic acid and benzylaminopurine. Auxin proved to be essential for somatic embryogenesis induction in nodal segments using 1 and 4 mg⋅L–1 of 2,4-dichlorophenoxyacetic acid alone. The results of this experiment will help advancing protocols for regeneration of somatic embryos and elucidating the physiological, molecular, and genetic mechanisms involved in the process of somatic embryogenesis for M. dubia. The development of an efficient protocol for in vitro clonal propagation of this species also lays the groundwork for further optimization of the system for genetic transformation.

Ultrasound Image Segmentation and Classification of Benign and Malignant Thyroid Nodules on the Basis of Deep Learning

This study aimed to investigate the effect of an image denoising algorithm based on weighted low-rank matrix restoration on thyroid nodule ultrasound images. A total of 1000 original ultrasound image data sets of thyroid nodules were selected as the study samples. The nodule segmentation data set of thyroid ultrasound region of interest (ROI) images was drawn and acquired. By introducing multiscale features and an attention mechanism to optimize the U-Net model, an ultrasound image segmentation model (F-U-Net) was constructed. The performance of the traditional U network model and full convolutional neural network model (FCN) was analyzed and compared by simulation experiments. The results showed that the dice coefficient, accuracy, and recall of the improved loss function in this study were significantly higher than those of the traditional cross entropy loss function and dice coefficient loss function, and the differences were statistically significant (P < 0.05). The Dice coefficient, accuracy, and recall of the F-U-net model were significantly higher than those of the traditional FCN model and U-net model (P < 0.05). The diagnostic sensitivity, specificity, accuracy, and positive predictive value of the F-U-net model for benign and malignant thyroid nodules were significantly higher than those of the FCN model and U-net model (P < 0.05). In summary, the proposed F-U network can effectively process the ultrasound images of thyroid nodules, improve the image quality, and help to improve the diagnostic effect of benign and malignant thyroid nodules. It provides a data reference for segmentation and reconstruction of benign and malignant ultrasound images of thyroid nodules.

An improved SPH method for simulating crack propagation and coalescence in rocks with pre-existing cracks

An improved smoothed particle hydrodynamics (SPH) method is proposed to simulate the failure process of rock samples with pre-existing cracks under compression loads. By improving the kernel function of SPH, the brittle fracture characteristics of rock materials are captured. In the improved SPH, a prefabricated node segment (PNS) method is used to generate arbitrarily complex open or closed cracks, avoiding the kernel truncation caused by digging out particles. In order to improve numerical stability, the tension instability control (TIC) technique is embedded in the momentum equation. Moreover, the improved SPH adopts the coupled dynamic buffer solid boundary treatment algorithm, which reduces the configuration of boundary virtual particles, makes the application of boundary conditions more efficient. The accuracy of the improved SPH method and the feasibility of TIC technique are verified by a benchmark example. Then, the effect of the center dislocation spacing of the pre-existing cross crack on the failure mode and compressive strength of rock samples is studied. In addition, the effects of confining pressure on the coalescence type and penetration failure mode of cracks in a rock sample with two pre-existing cracks are also investigated. The numerical results agree well with the experimental results, which verifies the accuracy and validity of the improved SPH method.

Deep Fuzzy SegNet-based lung nodule segmentation and optimized deep learning for lung cancer detection

Deep Fuzzy SegNet-based lung nodule segmentation and optimized deep learning for lung cancer detection

Efficient Procedure for Induction Somatic Embryogenesis in Holm Oak: Roles of Explant Type, Auxin Type, and Exposure Duration to Auxin

Holm oak is the dominant tree species in the Mediterranean climate. Currently, worrisome degradation of its ecosystems has been observed, produced, among other factors, by changes in land use, extreme weather events, forest fires, climate change, and especially the increasingly frequent episodes of high tree mortality caused by “oak decline”, which has brought with it a social concern that transcends the productive interest. Breeding and conservation programs for this species are necessary to ensure the prevalence of these ecosystems for future generations. Biotechnological tools such as somatic embryogenesis (SE) have great potential value for tree improvement and have been shown to be highly efficient in the propagation and conservation of woody species. One challenge to this approach is that SE induction in holm oak has not yet been optimized. Here, we present a new reproducible procedure to induce SE in holm oak; we evaluated the responsiveness of different initial explants exposed to different types, concentrations, and durations of auxin. SE rates were significantly improved (37%) by culturing nodal segments for two weeks in induction medium. In addition, a significant auxin–genotype interaction was observed.

Thidiazuron Induced In Vitro Plant Regeneration, Phenolic Contents, Antioxidant Potential, GC-MS Profiles and Nuclear Genome Stability of Plectranthus amboinicus (Lour.) Spreng

Plectranthus amboinicus (Lour.) Spreng is a perennial plant from the mint family with aromatic, succulent leaves and several health benefits. Multiple shoot regeneration was accomplished in vitro using nodal segments (NS) explants of P. amboinicus pretreated with 0, 0.5, 5, 25, 50, and 100 μM thidiazuron (TDZ) for 4 h, then transferred to a growth regulator-free media. After 8 weeks of growth, NS explants pre-treated with 25 μM TDZ for 4 h and then transferred to TDZ-free Murashige and Skoog (MS) media produced the greatest number of shoots (27.3 per NS) with the longest average shoot length (4.9 mm) in 97.2% of cultures. On the same medium, regeneration of roots in most of the P. amboinicus shoots occurred spontaneously. The in vitro-regenerated P. amboinicus plantlets were adequately hardened off and adapted to the ex-vitro environment with a 90% survival rate. Total phenolic, tannin, and flavonoid contents, as well as 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical-scavenging/antioxidant activity, were significantly higher in in vitro-regenerated plants than in ex vitro-plants. Flow cytometry (FCM) analysis validated the nuclear genome stability of the in vitro generated plants, which assessed their nuclear DNA content and found it to be comparable in genome size to that of the field-grown plants. The study found a quick and efficient method for in vitro multiplication of P. amboinicus which can aid to increased availability and accessibility of this plant species for various purposes. The genetic and phytochemical analysis of the in vitro propagated plants can also provide valuable insight into the plant’s properties and potential applications, which can further assist in its preservation and sustainable usage.

Accurate segmentation of lung nodule with low contrast boundaries by least weight navigation

Accurate segmentation of lung nodule with low contrast boundaries by least weight navigation

In vitro regeneration through organogenesis in Costus igneus–an important herbal insulin plant

Costus igneus is an important high-valued herbal insulin plant. An efficient protocol for in vitro plant regeneration through organogenesis from leaf and nodal explants was standardized. Explants were incubated in Murashige and Skoog (MS) medium amended with diverse levels of plant growth regulators (PGRs). Optimum frequency (68.25%) of adventitious shoot regeneration was obtained from nodal segments cultured onto MS medium containing 2.0 mg/l of 6-benzylaminopurine (BAP) with an average number of 5.0 shoots per explant. Further, MS medium containing 1.5 mg/l 2, 4- dichlorophenoxyacetic acid (2, 4-D) and 1.0 mg/l Thidiazuron (TDZ) produced a higher frequency of callus induction (55.22%) from leaf explants. Optimum shoot regeneration response (66.0%) through indirect organogenesis was observed from leaf-derived callus on MS medium fortified with BAP (2.0 mg/l) and kinetin (0.5 mg/l) with 6 shoots per callus with mean shoot length of 8.0 cm. The shootlets obtained from node and leaf cultures were rooted with a frequency of (63.52%) on MS medium with 1.0 mg/l indole-3-acetic acid (IAA). Plantlets obtained through organogenesis were successfully acclimatized in the greenhouse and field conditions. There were no significant differences among frequency of somatic plants regenerated through either direct or indirect organogenesis. The present study reports on successful plant regeneration through direct and indirect organogenesis in C. igneus. The regeneration protocols discussed here can be used effectively for large-scale multiplication, germplasm survival, pharmacological and genetic manipulation research.

DEHA-Net: A Dual-Encoder-Based Hard Attention Network with an Adaptive ROI Mechanism for Lung Nodule Segmentation.

Measuring pulmonary nodules accurately can help the early diagnosis of lung cancer, which can increase the survival rate among patients. Numerous techniques for lung nodule segmentation have been developed; however, most of them either rely on the 3D volumetric region of interest (VOI) input by radiologists or use the 2D fixed region of interest (ROI) for all the slices of computed tomography (CT) scan. These methods only consider the presence of nodules within the given VOI, which limits the networks' ability to detect nodules outside the VOI and can also encompass unnecessary structures in the VOI, leading to potentially inaccurate segmentation. In this work, we propose a novel approach for 3D lung nodule segmentation that utilizes the 2D region of interest (ROI) inputted from a radiologist or computer-aided detection (CADe) system. Concretely, we developed a two-stage lung nodule segmentation technique. Firstly, we designed a dual-encoder-based hard attention network (DEHA-Net) in which the full axial slice of thoracic computed tomography (CT) scan, along with an ROI mask, were considered as input to segment the lung nodule in the given slice. The output of DEHA-Net, the segmentation mask of the lung nodule, was inputted to the adaptive region of interest (A-ROI) algorithm to automatically generate the ROI masks for the surrounding slices, which eliminated the need for any further inputs from radiologists. After extracting the segmentation along the axial axis, at the second stage, we further investigated the lung nodule along sagittal and coronal views by employing DEHA-Net. All the estimated masks were inputted into the consensus module to obtain the final volumetric segmentation of the nodule. The proposed scheme was rigorously evaluated on the lung image database consortium and image database resource initiative (LIDC/IDRI) dataset, and an extensive analysis of the results was performed. The quantitative analysis showed that the proposed method not only improved the existing state-of-the-art methods in terms of dice score but also showed significant robustness against different types, shapes, and dimensions of the lung nodules. The proposed framework achieved the average dice score, sensitivity, and positive predictive value of 87.91%, 90.84%, and 89.56%, respectively.

Fusing enhanced Transformer and large kernel CNN for malignant thyroid nodule segmentation

Quick and accurate diagnosis for malignant thyroid nodules via ultrasonography is a valuable but challenging task even for experienced radiologists because of the complexity and variability of ultrasound images. Many computer-aided diagnosis (CAD) methods have been proposed to assist radiologists by providing objective suggestions. However, most existing segmentation approaches lack the ability to keep precise shape information and capture global long-range dependencies. To settle the above issues, we propose a deep learning-based CAD method called Transformer fusing CNN Network (TCNet) to segment malignant thyroid nodules automatically. Our proposed TCNet contains a large kernel CNN branch and an enhanced Transformer branch. In the former branch, we devise a Large Kernel Module (LKM) to extract the precise shape features of malignant thyroid nodules in ultrasound images. While in the latter branch, we design an Enhanced Transformer Module (ETM) to establish the remote connection between thyroid nodule pixels. To integrate multiscale feature maps produced from different phases of both branches, we develop a Multiscale Fusion Module (MFM) to connect the two branches. We compare the proposed model with several current commonly used segmentation methods on the MTNS dataset and other public datasets. The experimental results demonstrate the superiority and effectiveness of our method.

A Novel Model of Thyroid Nodule Segmentation for Ultrasound Images.

Ultrasonography is regarded as an effective technique for the detection, diagnosis and monitoring of thyroid nodules. Segmentation of thyroid nodules on ultrasound images is important in clinical practice. However, because in ultrasound images there is an unclear boundary between thyroid nodules and surrounding tissues, the accuracy of segmentation remains a challenge. Although the deep learning model provides an accurate and convenient method for thyroid nodule segmentation, it is unsatisfactory of the existing model in segmenting the margin of thyroid nodules. In this study, we developed boundary attention transformer net (BTNet), a novel segmentation network with a boundary attention mechanism combining the advantages of a convolutional neural network and transformer, which could fuse the features of both long and short ranges. Boundary attention is improved to focus on learning the boundary information, and this module enhances the segmentation ability of the network boundary. For features of different scales, we also incorporate a deep supervision mechanism to blend the outputs of different levels to enhance the segmentation effect. As the BTNet model incorporates the long range-short range connectivity effect and the boundary-regional cooperation capability, our model has excellent segmentation performance in thyroid nodule segmentation. The development of BTNet was based on the data set from Shanghai Jiao Tong University School of Medicine Affiliated Sixth People's Hospital and the public data set. BTNet achieved good performance in the segmentation of thyroid nodules with an intersection-over-union of 0.810 and Dice coefficient of 0.892 Moreover, our work revealed great improvement in the boundary metrics; for example, the boundary distance was 7.308, the boundary overlap 0.201 and the boundary Dice 0.194, all with p values <0.05.

Improving GAN Learning Dynamics for Thyroid Nodule Segmentation.

Thyroid nodules are lesions requiring diagnosis and follow-up. Tools for detecting and segmenting nodules can help physicians with this diagnosis. Besides immediate diagnosis, automated tools can also enable tracking of the probability of malignancy over time. This paper demonstrates a new algorithm for segmenting thyroid nodules in ultrasound images. The algorithm combines traditional supervised semantic segmentation with unsupervised learning using GANs. The hybrid approach has the potential to upgrade the semantic segmentation model's performance, but GANs have the well-known problems of unstable learning and mode collapse. To stabilize the training of the GAN model, we introduce the concept of closed-loop control of the gain on the loss output of the discriminator. We find gain control leads to smoother generator training and avoids the mode collapse that typically occurs when the discriminator learns too quickly relative to the generator. We also find that the combination of the supervised and unsupervised learning styles encourages both low-level accuracy and high-level consistency. As a test of the concept of controlled hybrid supervised and unsupervised semantic segmentation, we introduce a new model named the StableSeg GAN. The model uses DeeplabV3+ as the generator, Resnet18 as the discriminator, and uses PID control to stabilize the GAN learning process. The performance of the new model in terms of IoU is better than DeeplabV3+, with mean IoU of 81.26% on a challenging test set. The results of our thyroid nodule segmentation experiments show that StableSeg GANs have flexibility to segment nodules more accurately than either comparable supervised segmentation models or uncontrolled GANs.