Automatic summarization of endoscopic skull base surgical videos through object detection and hidden Markov modeling.

Endoscopic endonasal surgery is a medical procedure that utilizes an endoscopic video camera to view and manipulate a surgical site accessed through the nose. Despite these surgeries being video recorded, these videos are seldom reviewed or even saved in patient files due to the size and length of the video file. Editing to a manageable size may necessitate viewing 3 h or more of surgical video and manually splicing together the desired segments. We suggest a novel multi-stage video summarization procedure utilizing deep semantic features, tool detections, and video frame temporal correspondences to create a representative summarization. Summarization by our method resulted in a 98.2% reduction in overall video length while preserving 84% of key medical scenes. Furthermore, resulting summaries contained only 1% of scenes with irrelevant detail such as endoscope lens cleaning, blurry frames, or frames external to the patient. This outperformed leading commercial and open source summarization tools not designed for surgery, which only preserved 57% and 46% of key medical scenes in similar length summaries, and included 36% and 59% of scenes containing irrelevant detail. Experts agreed that on average (Likert Scale = 4) that the overall quality of the video was adequate to share with peers in its current state.

Melatonin inhibited the progression of gastric cancer induced by Bisphenol S via regulating the estrogen receptor 1.

In recent years, Bisphenol S (BPS) has increasingly been used as an alternative to Bisphenol A (BPA) in food, paper, and personal care products. It is imperative to clarify the relationship between BPS and tumors in order to treat and prevent diseases. This study discovered a new method for predicting tumor correlations between BPS interactive genes. According to analyses conducted by Gene Ontology and Kyoto Encyclopedia of Genes and Genomes, interactive genes were primarily found in gastric cancer. Based on gene-targeted prediction and molecular docking, BPS appears to exert potential gastric cancer-causing effects through estrogen receptor 1 (ESR1). In addition, gastric cancer patients' prognosis could be accurately predicted by a bisphenol-based prognostic prediction model. Subsequently, the proliferation and migration abilities of gastric cancer cells were further demonstrated to be significantly enhanced by BPS. Similarly, molecular docking analysis revealed that melatonin is also highly correlated with gastric cancer and BPS. In cell proliferation and migration assays, melatonin and BPS exposure inhibited the invasion abilities of gastric cancer cells compared to BPS-exposure. Our research provided a new direction for the exploration the correlation between cancer and environmental toxicity.

Deep causal learning for robotic intelligence.

This invited Review discusses causal learning in the context of robotic intelligence. The Review introduces the psychological findings on causal learning in human cognition, as well as the traditional statistical solutions for causal discovery and causal inference. Additionally, we examine recent deep causal learning algorithms, with a focus on their architectures and the benefits of using deep nets, and discuss the gap between deep causal learning and the needs of robotic intelligence.

Nomogram for prediction of prolonged postoperative ileus after colorectal resection.

BACKGROUND: Prolonged postoperative ileus (PPOI) is a major complication in patients undergoing colorectal resection. The aim of this study was to analyze the risk factors contributing to PPOI, and to develop an effective nomogram to determine the risks of this population. METHODS: A total of 1,254 patients with colorectal cancer who underwent radical colorectal resection at Fujian Cancer Hospital from March 2016 to August 2021 were enrolled as a training cohort in this study. Univariate analysis and multivariate logistic regressions were performed to determine the correlation between PPOI and clinicopathological characteristics. A nomogram predicting the incidence of PPOI was constructed. The cohort of 153 patients from Fujian Provincial Hospital were enrolled as a validation cohort. Internal and external validations were used to evaluate the prediction ability by area under the receiver operating characteristic curve (AUC) and a calibration plot. RESULTS: In the training cohort, 128 patients (10.2%) had PPOI after colorectal resection. The independent predictive factors of PPOI were identified, and included gender, age, surgical approach and intraoperative fluid overload. The AUC of nomogram were 0.779 (95% CI: 0.736-0.822) and 0.791 (95%CI: 0.677-0.905) in the training and validation cohort, respectively. The two cohorts of calibration plots showed a good consistency between nomogram prediction and actual observation. CONCLUSIONS: A highly accurate nomogram was developed and validated in this study, which can be used to provide individual prediction of PPOI in patients after colorectal resection, and this predictive power can potentially assist surgeons to make the optimal treatment decisions.

iGRLCDA: identifying circRNA-disease association based on graph representation learning.

While the technologies of ribonucleic acid-sequence (RNA-seq) and transcript assembly analysis have continued to improve, a novel topology of RNA transcript was uncovered in the last decade and is called circular RNA (circRNA). Recently, researchers have revealed that they compete with messenger RNA (mRNA) and long noncoding for combining with microRNA in gene regulation. Therefore, circRNA was assumed to be associated with complex disease and discovering the relationship between them would contribute to medical research. However, the work of identifying the association between circRNA and disease in vitro takes a long time and usually without direction. During these years, more and more associations were verified by experiments. Hence, we proposed a computational method named identifying circRNA-disease association based on graph representation learning (iGRLCDA) for the prediction of the potential association of circRNA and disease, which utilized a deep learning model of graph convolution network (GCN) and graph factorization (GF). In detail, iGRLCDA first derived the hidden feature of known associations between circRNA and disease using the Gaussian interaction profile (GIP) kernel combined with disease semantic information to form a numeric descriptor. After that, it further used the deep learning model of GCN and GF to extract hidden features from the descriptor. Finally, the random forest classifier is introduced to identify the potential circRNA-disease association. The five-fold cross-validation of iGRLCDA shows strong competitiveness in comparison with other excellent prediction models at the gold standard data and achieved an average area under the receiver operating characteristic curve of 0.9289 and an area under the precision-recall curve of 0.9377. On reviewing the prediction results from the relevant literature, 22 of the top 30 predicted circRNA-disease associations were noted in recent published papers. These exceptional results make us believe that iGRLCDA can provide reliable circRNA-disease associations for medical research and reduce the blindness of wet-lab experiments.

Real-time virtual intraoperative CT in endoscopic sinus surgery.

PURPOSE: Endoscopic sinus surgery (ESS) is typically guided under preoperative computed tomography (CT), which increasingly diverges from actual patient anatomy as the surgery progresses. Studies have reported that the revision surgery rate in ESS ranges between 28 and 47%. This paper presents a method that can update the preoperative CT in real time to improve surgical completeness in ESS. APPROACH: The work presents and compares three novel methods that use instrument motion data and anatomical structures to predict surgical modifications in real time. The methods use learning techniques, such as nonparametric filtering and Gaussian process regression, to correlate surgical modifications with instrument tip positions, tip trajectories, and instrument shapes. Preoperative CT image sets are updated with modification predictions to serve as a virtual intraoperative CT. RESULTS: The three methods were compared in eight ESS cadaver cases, which were performed by five surgeons and included the following representative ESS operations: maxillary antrostomy, uncinectomy, anterior and posterior ethmoidectomy, and sphenoidotomy. Experimental results showed accuracy metrics that were clinically acceptable with dice similarity coefficients > 86%, with F-score > 92% and precision > 89.91% in surgical completeness evaluation. Among the three methods, the tip trajectory-based estimator had the highest precision of 96.87%. CONCLUSIONS: This work demonstrated that virtually modified intraoperative CT scans improved the consistency between the actual surgical scene and the reference model, and could lead to improved surgical completeness in ESS. Compared to actual intraoperative CT scans, the proposed method has no impact on existing surgical protocols, does not require extra hardware, does not expose the patient to radiation, and does not lengthen time under anesthesia.

Under-reporting of COVID-19 in the Northern Health Authority region of British Columbia.

Asymptomatic and pauci-symptomatic presentations of COVID-19 along with restrictive testing protocols result in undetected COVID-19 cases. Estimating undetected cases is crucial to understanding the true severity of the outbreak. We introduce a new hierarchical disease dynamics model based on the N-mixtures hidden population framework. The new models make use of three sets of disease count data per region: reported cases, recoveries and deaths. Treating the first two as under-counted through binomial thinning, we model the true population state at each time point by partitioning the diseased population into the active, recovered and died categories. Both domestic spread and imported cases are considered. These models are applied to estimate the level of under-reporting of COVID-19 in the Northern Health Authority region of British Columbia, Canada, during 30 weeks of the provincial recovery plan. Parameter covariates are easily implemented and used to improve model estimates. We compare two distinct methods of model-fitting for this case study: (1) maximum likelihood estimation, and (2) Bayesian Markov chain Monte Carlo. The two methods agreed exactly in their estimates of under-reporting rate. When accounting for changes in weekly testing volumes, we found under-reporting rates varying from 60.2% to 84.2%.

Le recours à des protocoles de tests restrictifs et l'existence de formes asymptomatiques et paucisymptomatiques de la COVID­19 contribuent à la non détection de cas COVID­19. Pour comprendre la véritable gravité de l'épidémie, il est primordial d'estimer correctement le nombre de cas non détectés. A cette fin, les auteurs de ce travail proposent un nouveau modèle hiérarchique des dynamiques de la maladie basé sur l'approche de N­mélanges de population cachée. Ces modèles utilisent trois types de données régionales, à savoir, les nombres de cas déclarés, guéris et décédés. En faisant appel à l'amincissement binomial (binomial thinning) et en traitant les nombres de cas déclarés et guéris comme étant sous­évalués, les auteurs proposent une modélisation de l'état réel de l'épidémie basée sur une partition de la population malade en trois catégories : cas actifs, cas guéris et cas décédés. Cette partition tient compte des cas de propagation intérieure et des cas importés. Les auteurs ont utilisé les données recueillies durant les trente semaines du plan de rétablissement provincial de la région de l'Autorité sanitaire du Nord de la Colombie­Britannique, Canada pour illustrer leur approche et estimer le niveau de sous­déclaration COVID­19 associé. Des covariables peuvent être facilement incorporées au modèle proposé et améliorer la qualité des estimations. Deux méthodes d'ajustement sont retenues: (1) l'estimation par maximum de vraisemblance, et (2) la méthode de Monte Carlo par chaînes de Markov. Les estimations du taux de sous­déclaration obtenues par ces deux méthodes concordent exactement et varient entre 60,2% et 84,2% après ajustement des variations des volumes de tests hebdomadaires.

Interdigitating dendritic cell sarcoma located in the adrenal gland: a case report and literature review.

We report a case of interdigitating dendritic cell sarcoma (IDCS) originating from the adrenal gland. A 57-year-old middle-aged woman with no previous history of malignancy came to our hospital after color Doppler ultrasound revealed a right adrenal mass. An abdominal computed tomography scan also showed an adrenal mass. Postoperative pathology confirmed the diagnosis of IDCS. After complete surgical removal of the adrenal tumor, the patient has been disease-free for 1 year. IDCS may have a good prognosis after surgical resection. To our knowledge, this is only the second reported case of IDCS in the adrenal region.

Application of Primary/Secondary Circulating Tumor Cells for the Prediction of Biochemical Recurrence in Nonmetastatic Prostate Cancer Patients following Radical Prostatectomy or Radiotherapy: A Meta-Analysis.

BACKGROUND: Circulating tumor cells (CTCs) have been regarded as an independent prognostic marker for metastatic castration-resistant prostate cancer (mCRPC). Its prognostic value, however, in nonmetastatic prostate cancer (NMPC) is still unclear. PURPOSE: To elucidate whether CTCs can predict the biochemical recurrence (BCR) in NMPC patients following radical prostatectomy (RP) or radiotherapy (RT). METHODS: PubMed, Cochrane Database, and Embase and the references in relevant studies were systematically searched. Studies that investigated the correlation of CTCs and BCR in NMPC patients after RP or RT were identified and reviewed. Overall odds ratio (OR) of BCR in such patients with/without CTCs was pooled. We also calculated and pooled overall prevalence of BCR in such CTC-positive patients. RESULTS: In total, 12 studies comprising 1917 participants were eligible for the meta-analysis and showed that the presence of secondary circulating tumor cells (SCTCs) is associated with a higher BCR rate of 59% (95% CI: 22%-88%) in patients with NMPC after RP or RT (OR = 6.12; 95% CI: 2.22-16.85; P < 0.001). However, regardless of the presence of primary circulating tumor cells (PCTCs), it has not been shown to be associated with higher BCR. CONCLUSIONS: Our research demonstrated that SCTC-positive patients are associated with higher BCR compared to SCTC-negative patients in NMPC. Therefore, it is recommended that NMPC patients undergo CTC surveillance intensively after RP or RT.

Total Synthesis of Dysifragilones A and B and Dysidavarone C.

The concise synthesis of dysifragilones A and B and dysidavarones has been accomplished for the first time in a divergent way from a common intermediate. The synthetic route features an intramolecular reductive Heck reaction to construct the 6/5/6/6/-tetracycle of dysifragilones A and B and an intramolecular palladium-catalyzed α-arylation of a sterically hindered ketone to forge the tetracyclo[7.7.1.02,7.010,15]heptadecane core structure of dysidavarone C. The late-stage introduction of amino and ethoxy groups is effective.

γ-Synuclein is Closely Involved in Autophagy that Protects Colon Cancer Cell from Endoplasmic Reticulum Stress.

BACKGROUND: In previous studies, we provided evidence suggesting the involvement of γ-synuclein in growth, invasion, and metastasis of colon cancer cells in vitro and in vivo. Among γ-synuclein downstream genes, the microtubule-associated protein 1 Light Chain 3 (LC3), an autophagy gene, was screened by gene expression profile chip analysis. OBJECTIVE: We planned to investigate the functional effects of γ-synuclein on autophagy induced by ER stress in colon cancer cells. METHODS: We investigated the functional effects of γ-synuclein on autophagy and apoptosis induced by Thapsigargin (TG), ER stress-inducing agent, in colon cancer cell lines using immunofluorescence staining, RT-PCR, western blot, CCK8 test, flow cytometry analysis, and transmission electron microscopy. To further determine how γ-synuclein regulated autophagy and apoptosis, PD98059 (ERK inhibitor), SP600125 (ERK inhibitor), anisomycin (JNK activator), and c-Jun siRNA were used respectively in γ-synuclein siRNA transfected HCT116 cells. Then, autophagy proteins, apoptosis proteins, and pathway proteins were detected by western blot analysis. The expression of autophagy genes was assessed by RT-PCR. RESULTS: Our data showed that ER stress-induced colon cancer cells autophagy mainly in the early stage (0-24h) and apoptosis mainly in the late stage (24-48h). ER stress up-regulated γ-synuclein gene and protein expression in colon cancer cells, accompanied by autophagy. γ-synuclein protected HCT116 cells by enhancing autophagy in the early stage (0-24h) through activation of ERK and JNK pathway and inhibiting apoptosis in the late stage (24-48h) through inhibition of the JNK pathway. γ-synuclein could promote autophagy via the JNK pathway activation of ATG genes, LC3, Beclin 1, and ATG7. γ-synuclein may play a role in the transition between autophagy and apoptosis in our model. CONCLUSION: Overall, we provided the first experimental evidence to show that γ-synuclein may play an important role in autophagy that protects colon cancer cells from ER stress. Therefore, our data suggest a new molecular mechanism for γ-synuclein-mediated CRC progression.

Dense Depth Estimation from Stereo Endoscopy Videos Using Unsupervised Optical Flow Methods.

In the context of Minimally Invasive Surgery, estimating depth from stereo endoscopy plays a crucial role in three-dimensional (3D) reconstruction, surgical navigation, and augmentation reality (AR) visualization. However, the challenges associated with this task are three-fold: 1) feature-less surface representations, often polluted by artifacts, pose difficulty in identifying correspondence; 2) ground truth depth is difficult to estimate; and 3) an endoscopy image acquisition accompanied by accurately calibrated camera parameters is rare, as the camera is often adjusted during an intervention. To address these difficulties, we propose an unsupervised depth estimation framework (END-flow) based on an unsupervised optical flow network trained on un-rectified binocular videos without calibrated camera parameters. The proposed END-flow architecture is compared with traditional stereo matching, self-supervised depth estimation, unsupervised optical flow, and supervised methods implemented on the Stereo Correspondence and Reconstruction of Endoscopic Data (SCARED) Challenge dataset. Experimental results show that our method outperforms several state-of-the-art techniques and achieves a close performance to that of supervised methods.

CILP2 overexpression correlates with tumor progression and poor prognosis in patients with colorectal cancer in The Cancer Genome Atlas (TCGA) study.

BACKGROUND: Genetic alterations play an important role in the progression of colorectal cancer (CRC). Identifying new biomarkers to assess the prognosis of patients with CRC is critical. Cartilage intermediate layer protein 2 (CILP2) gene, screened from TCGA database by bioinformatics, may be closely related to the progression of CRC. CILP2 was barely reported with clinical features of tumors. MATERIALS AND METHODS: Clinical information and RNA-seq data were derived from TCGA colorectal carcinoma cohort. CILP2 expression at mRNA level was estimated by bioinformatical analysis of TCGA cases. Tissue microarray (TMA) was constructed containing paraffin-embedded 64 pairs of CRC and matched adjacent normal tissues. The expression at the protein level was detected in 64 pairs of CRC and matched adjacent normal tissues by immunohistochemical analysis. CILP2 expression level and its clinical value were estimated by bioinformatical analysis with linear and logistic regression. Survival analysis was performed between high and low groups of CILP2 expression by Cox regression analysis, and the P value was calculated by the log-rank test. The Kaplan-Meier curves were tested by the log-rank test. RESULTS: CILP2 was statistically significantly higher expressed in the CRC tissues when compared with paired adjacent normal tissues in TCGA cohort (P < 0.001) and in the TMA cohort (P = 0.001). Also, CILP2 high expression was strongly correlated with T3/4 stage (P = 0.001), N1/2/3 stage (P = 0.005), M1 stage (P = 0.048), and higher clinical stage (UICC 2010 stage) (P < 0.001) in TCGA cohort, and also positively associated with T3/4 stage (P = 0.022) and higher clinical stage (UICC 2010 stage) (P = 0.03) in TMA cohort. Furthermore, CILP2 overexpression predicted poor prognosis and could be an independent prognostic factor (P = 0.003). CONCLUSION: We revealed that CILP2 is associated with advanced stages and could play a role as an independent predictor of poor survival in CRC.

SPRY4-IT1 promotes survival of colorectal cancer cells through regulating PDK1-mediated glycolysis.

Colorectal cancer (CRC) becomes the third leading cause of cancer-related deaths worldwide recently. The prognosis of CRC is still poor in decades, and targeted therapy is still a potential effective treatment. Long non-coding RNAs (lncRNAs) could regulate series of cellular functions and developmental processes. LncRNA-SPRY4-IT1 (GenBank ID AK024556) is derived from an intron of the SPRY4 gene, which was highly expressed in melanoma cells and affected the progression of multiple types of cancers. However, the mechanism of SPRY4-IT1 in CRC progression remains unclear. Herein, we found the high level of SPRY4-IT1 in human colorectal cancer (CRC) tissues and cells, and correlated with patients' prognosis. We further noticed that SPRY4-IT1 regulated CRC cell growth and glycolysis, and promoting PDK1 expression. Our data further confirmed that SPRY4-IT1 regulated CRC progression targeting PDK1. We therefore thought SPRY4-IT1 could serve as a promising molecular target for the treatment of CRC.

Learning Representations to Predict Intermolecular Interactions on Large-Scale Heterogeneous Molecular Association Network.

Molecular components that are functionally interdependent in human cells constitute molecular association networks. Disease can be caused by disturbance of multiple molecular interactions. New biomolecular regulatory mechanisms can be revealed by discovering new biomolecular interactions. To this end, a heterogeneous molecular association network is formed by systematically integrating comprehensive associations between miRNAs, lncRNAs, circRNAs, mRNAs, proteins, drugs, microbes, and complex diseases. We propose a machine learning method for predicting intermolecular interactions, named MMI-Pred. More specifically, a network embedding model is developed to fully exploit the network behavior of biomolecules, and attribute features are also calculated. Then, these discriminative features are combined to train a random forest classifier to predict intermolecular interactions. MMI-Pred achieves an outstanding performance of 93.50% accuracy in hybrid associations prediction under 5-fold cross-validation. This work provides systematic landscape and machine learning method to model and infer complex associations between various biological components.

GCNCDA: A new method for predicting circRNA-disease associations based on Graph Convolutional Network Algorithm.

Numerous evidences indicate that Circular RNAs (circRNAs) are widely involved in the occurrence and development of diseases. Identifying the association between circRNAs and diseases plays a crucial role in exploring the pathogenesis of complex diseases and improving the diagnosis and treatment of diseases. However, due to the complex mechanisms between circRNAs and diseases, it is expensive and time-consuming to discover the new circRNA-disease associations by biological experiment. Therefore, there is increasingly urgent need for utilizing the computational methods to predict novel circRNA-disease associations. In this study, we propose a computational method called GCNCDA based on the deep learning Fast learning with Graph Convolutional Networks (FastGCN) algorithm to predict the potential disease-associated circRNAs. Specifically, the method first forms the unified descriptor by fusing disease semantic similarity information, disease and circRNA Gaussian Interaction Profile (GIP) kernel similarity information based on known circRNA-disease associations. The FastGCN algorithm is then used to objectively extract the high-level features contained in the fusion descriptor. Finally, the new circRNA-disease associations are accurately predicted by the Forest by Penalizing Attributes (Forest PA) classifier. The 5-fold cross-validation experiment of GCNCDA achieved 91.2% accuracy with 92.78% sensitivity at the AUC of 90.90% on circR2Disease benchmark dataset. In comparison with different classifier models, feature extraction models and other state-of-the-art methods, GCNCDA shows strong competitiveness. Furthermore, we conducted case study experiments on diseases including breast cancer, glioma and colorectal cancer. The results showed that 16, 15 and 17 of the top 20 candidate circRNAs with the highest prediction scores were respectively confirmed by relevant literature and databases. These results suggest that GCNCDA can effectively predict potential circRNA-disease associations and provide highly credible candidates for biological experiments.

A Model-Based Recurrent Neural Network With Randomness for Efficient Control With Applications.

Recently, Recurrent Neural Network (RNN) control schemes for redundant manipulators have been extensively studied. These control schemes demonstrate superior computational efficiency, control precision, and control robustness. However, they lack planning completeness. This paper explains why RNN control schemes suffer from the problem. Based on the analysis, this work presents a new random RNN control scheme, which 1) introduces randomness into RNN to address the planning completeness problem, 2) improves control precision with a new optimization target, 3) improves planning efficiency through learning from exploration. Theoretical analyses are used to prove the global stability, the planning completeness, and the computational complexity of the proposed method. Software simulation is provided to demonstrate the improved robustness against noise, the planning completeness and the improved planning efficiency of the proposed method over benchmark RNN control schemes. Real-world experiments are presented to demonstrate the application of the proposed method.

LMTRDA: Using logistic model tree to predict MiRNA-disease associations by fusing multi-source information of sequences and similarities.

Emerging evidence has shown microRNAs (miRNAs) play an important role in human disease research. Identifying potential association among them is significant for the development of pathology, diagnose and therapy. However, only a tiny portion of all miRNA-disease pairs in the current datasets are experimentally validated. This prompts the development of high-precision computational methods to predict real interaction pairs. In this paper, we propose a new model of Logistic Model Tree for predicting miRNA-Disease Association (LMTRDA) by fusing multi-source information including miRNA sequences, miRNA functional similarity, disease semantic similarity, and known miRNA-disease associations. In particular, we introduce miRNA sequence information and extract its features using natural language processing technique for the first time in the miRNA-disease prediction model. In the cross-validation experiment, LMTRDA obtained 90.51% prediction accuracy with 92.55% sensitivity at the AUC of 90.54% on the HMDD V3.0 dataset. To further evaluate the performance of LMTRDA, we compared it with different classifier and feature descriptor models. In addition, we also validate the predictive ability of LMTRDA in human diseases including Breast Neoplasms, Breast Neoplasms and Lymphoma. As a result, 28, 27 and 26 out of the top 30 miRNAs associated with these diseases were verified by experiments in different kinds of case studies. These experimental results demonstrate that LMTRDA is a reliable model for predicting the association among miRNAs and diseases.

An Improved Deep Forest Model for Predicting Self-Interacting Proteins From Protein Sequence Using Wavelet Transformation.

Self-interacting proteins (SIPs), whose more than two identities can interact with each other, play significant roles in the understanding of cellular process and cell functions. Although a number of experimental methods have been designed to detect the SIPs, they remain to be extremely time-consuming, expensive, and challenging even nowadays. Therefore, there is an urgent need to develop the computational methods for predicting SIPs. In this study, we propose a deep forest based predictor for accurate prediction of SIPs using protein sequence information. More specifically, a novel feature representation method, which integrate position-specific scoring matrix (PSSM) with wavelet transform, is introduced. To evaluate the performance of the proposed method, cross-validation tests are performed on two widely used benchmark datasets. The experimental results show that the proposed model achieved high accuracies of 95.43 and 93.65% on human and yeast datasets, respectively. The AUC value for evaluating the performance of the proposed method was also reported. The AUC value for yeast and human datasets are 0.9203 and 0.9586, respectively. To further show the advantage of the proposed method, it is compared with several existing methods. The results demonstrate that the proposed model is better than other SIPs prediction methods. This work can offer an effective architecture to biologists in detecting new SIPs.

Comparison of Micro-Computed Tomography and Clinical Computed Tomography Protocols for Visualization of Nasal Cartilage Before Surgical Planning for Rhinoplasty.

IMPORTANCE: There is no imaging standard to model nasal cartilage for the planning of rhinoplasty procedures. Preoperative visualization of cartilage may improve objective evaluation of nasal deformities, surgical planning, and surgical reconstruction. OBJECTIVES: To evaluate the feasibility of visualizing nasal cartilage using high resolution micro-computed tomography (CT) compared with the criterion standard of pathologic findings in a cadaveric specimen and to evaluate its accuracy compared with various clinical CT protocols. DESIGN, SETTING, AND PARTICIPANTS: Anatomic study at the University of Washington using single human cadaveric nasal specimens performed from July 10, 2017, to March 30, 2018. INTERVENTIONS: A micro-CT acquisition with 60-micron resolution was obtained of a nasal specimen. The specimen was then scanned with 5 different clinical CT protocols to span both clinical care and machine limits. The specimen was then sectioned in 5-mm axial slices for pathologic analysis. MAIN OUTCOMES AND MEASURES: Micro-CT images were registered to pathologic specimen cross-sections using a graphite fiducial system. Cartilage substructures were manually segmented and analyzed. A library of matched images across the micro-CT and various clinical CT protocols was then developed. Region of interest analysis was performed for each of the cartilage structures and their boundaries on clinical CT protocols and micro-CT, with the outcome of mean (SD) density using Hounsfield units. RESULTS: A single human cadaveric nasal specimen was used to obtain the following results. Lower lateral cartilage, upper lateral cartilage, and septal cartilage were accurately delineated on the micro-CT images compared with pathologic findings. The mean absolute deviation from pathologic findings was 0.30 mm for septal cartilage thickness, 0.98 mm for maximal upper lateral cartilage length, and 1.40 mm for maximal lower lateral cartilage length. On clinical CT protocols, only septal cartilage was well discriminated from boundary. Higher radiation dose resulted in more accurate density measurements of cartilage, but it did not ultimately improve ability to discriminate cartilage. CONCLUSIONS AND RELEVANCE: The results of this anatomic study may represent a notable step toward advancing knowledge of the capabilities and pitfalls of nasal cartilage visualization on CT. Nasal cartilage visualization was feasible on the micro-CT compared with pathologic findings. Future research may further examine the barriers to accurately visualizing upper lateral cartilage and lower lateral cartilage, a prerequisite for clinical application. LEVEL OF EVIDENCE: NA.