Comparative investigation of transport and deposition of nebulized particles in nasal airways following various middle turbinectomy

BACKGROUND

Topical intranasal medication is required following functional endoscopic sinus surgery (FESS). The optimal particle size of transnasal nebulization aimed at the sinonasal cavities is not conclusive. The current study aims to evaluate the effect of particle size and various surgery scope of middle turbinectomy (MT) on post-full FESS drug delivery to the sinonasal cavities.

METHODS

Sinonasal reconstructions were performed from post-full FESS CT scans in 6 chronic rhinosinusitis with nasal polyps (CRSwNP) patients. Four additional models representing alternative surgery scopes of MT were established from each post-FESS reconstruction for simulation data comparison. Airflow and particle deposition of nebulized delivery were simulated via computational fluid dynamics (CFD) and validated through in vitro experiments. The optimal particle sizes reaching a deposition of at least 75% of the maximum in the targeted regions were identified.

RESULTS

The drug deposition rate onto the targeted regions increased following MT, with the greatest deposition following posterior MT (P-MT). Droplets in the range of 18-26 μm reached a deposition of larger than 75% of the maximum onto the targeted regions. Drug delivery rate in the sinonasal cavities varied significantly among individuals and across different types of MT with varying surgical scopes.

CONCLUSIONS

This study is the first to investigate the effect of various surgery scope on drug delivery by transnasal nebulization to the sinonasal cavities. The findings strongly affirm the vast potential of transnasal nebulization as an effective post-FESS treatment option. Moreover, it emphasizes that the drug delivery process via atomizers to the nasal cavity and paranasal sinuses is highly sensitive to the particle size.

MAIRNet: weakly supervised anatomy-aware multimodal articulated image registration network

PURPOSE

Multimodal articulated image registration (MAIR) is a challenging problem because the resulting transformation needs to maintain rigidity for bony structures while allowing elastic deformation for surrounding soft tissues. Existing deep learning-based methods ignore the articulated structures and consider it as a pure deformable registration problem, leading to suboptimal results.

METHODS

We propose a novel weakly supervised anatomy-aware multimodal articulated image registration network, referred as MAIRNet, to solve the challenging problem. The architecture of MAIRNet comprises of two branches: a non-learnable polyrigid registration branch to estimate an initial velocity field, and a learnable deformable registration branch to learn an increment. These two branches work together to produce a velocity field that can be integrated to generate the final displacement field.

RESULTS

We designed and conducted comprehensive experiments on three datasets to evaluate the performance of the proposed method. Specifically, on the hip dataset, our method achieved, respectively, an average dice of 90.8%, 92.4% and 91.3% for the pelvis, the right femur, and the left femur. On the lumbar spinal dataset, our method obtained, respectively, an average dice of 86.1% and 85.9% for the L4 and the L5 vertebrae. On the thoracic spinal dataset, our method achieved, respectively, an average dice of 76.7%, 79.5%, 82.9%, 85.5% and 85.7% for the five thoracic vertebrae ranging from T6 to T10.

CONCLUSION

In summary, we developed a novel approach for multimodal articulated image registration. Comprehensive experiments conducted on three typical yet challenging datasets demonstrated the efficacy of the present approach. Our method achieved better results than the state-of-the-art approaches.

Ligamentum Teres Tears and Increased Combined Anteversion Are Associated With Hip Microinstability in Patients With Borderline Dysplasia

PURPOSE

To investigate the differences in the prevalence of ligamentum teres (LT) tears and other radiographic measurements in borderline dysplasia of the hip (BDDH) with/without microinstability and to evaluate the associations between these imaging findings and the prevalence of microinstability in patients with BDDH.

METHODS

This was a retrospective study of symptomatic patients with BDDH (18° ≤ lateral center-edge angle <25°) treated with arthroscopy in our hospital between January 2016 and December 2021. These patients were divided into the BDDH with microinstability (mBDDH) group and the stable BDDH (nBDDH) group. The radiographic parameters associated with hip joint stability, such as the state of LT, acetabular versions, femoral neck version, Tönnis angle, combined anteversions, and anterior/posterior acetabular coverage, were reviewed and analyzed.

RESULTS

There were 54 patients (49 female/5 male, 26.7 ± 6.9 years) in the mBDDH group and 81 patients (74 female/7 male, 27.2 ± 7.7 years) in the nBDDH group. The mBDDH group had greater LT tear (43/54 vs 5/81) and general laxity rates, increased femoral neck version, acetabular version and combined anteversion (52.4 ± 5.9 vs 41.5 ± 7.1 at 3-o'clock level) than the nBDDH group. Binary logistic regression showed that LT tears (odds ratio 6.32, 95% confidence interval 1.38-28.8; P = .02; R2 = .458) and combined anteversion at the 3-o'clock level (odds ratio 1.42, 95% confidence interval 1.09-1.84; P < .01; R2 = .458) were independent predictors of microinstability in patients with BDDH. The cutoff value of combined anteversion at 3-o'clock level was 49.5°. In addition, LT tear was correlated with increased combined anteversion at 3-o'clock level in patients with BDDH (P < .01, η2 = 0.29).

CONCLUSIONS

LT tears and increased combined anteversion at the 3-o'clock level on the acetabular clockface were associated with hip microinstability in patients with BDDH, suggesting that patients with BDDH and LT tears might have a greater prevalence of anterior microinstability.

LEVEL OF EVIDENCE

Level III, case‒control study.

SMILE: Siamese Multi-scale Interactive-representation LEarning for Hierarchical Diffeomorphic Deformable image registration

Deformable medical image registration plays an important role in many clinical applications. It aims to find a dense deformation field to establish point-wise correspondences between a pair of fixed and moving images. Recently, unsupervised deep learning-based registration methods have drawn more and more attention because of fast inference at testing stage. Despite remarkable progress, existing deep learning-based methods suffer from several limitations including: (a) they often overlook the explicit modeling of feature correspondences due to limited receptive fields; (b) the performance on image pairs with large spatial displacements is still limited since the dense deformation field is regressed from features learned by local convolutions; and (c) desirable properties, including topology-preservation and the invertibility of transformation, are often ignored. To address above limitations, we propose a novel Convolutional Neural Network (CNN) consisting of a Siamese Multi-scale Interactive-representation LEarning (SMILE) encoder and a Hierarchical Diffeomorphic Deformation (HDD) decoder. Specifically, the SMILE encoder aims for effective feature representation learning and spatial correspondence establishing while the HDD decoder seeks to regress the dense deformation field in a coarse-to-fine manner. We additionally propose a novel Local Invertible Loss (LIL) to encourage topology-preservation and local invertibility of the regressed transformation while keeping high registration accuracy. Extensive experiments conducted on two publicly available brain image datasets demonstrate the superiority of our method over the state-of-the-art (SOTA) approaches. Specifically, on the Neurite-OASIS dataset, our method achieved an average DSC of 0.815 and an average ASSD of 0.633 mm.

ELABELA-APJ Axis Enhances Mesenchymal Stem Cell Proliferation and Migration via the METTL3/PI3K/AKT Pathway

Mesenchymal stem cells (MSCs) possess a strong therapeutic potential in regenerative medicine. ELABELA (ELA) is a 32 amino acid peptide that binds to the apelin peptide jejunum receptor (APJ) to regulate cell proliferation and migration. The aim of this study was to investigate the function of ELA vis-a-vis the MSC proliferation and migration, and further explore the underlying mechanism. We demonstrated that the exogenous supplement of ELA boosts the proliferation and migration ability of MSCs, alongside improved in vitro cell viability. These capabilities were rendered moot upon APJ knockdown. In addition, ELA (5-20 μM) was shown to upregulate the expression of METTL3 in a concentrationdependent pattern, a capacity which was suppressed by APJ reduction, whereas the downregulation of METTL3 expression blocked the beneficial effects induced by ELA. ELA was also observed to upregulate the phosphorylation level of AKT. This ELA-induced activation of the PI3K/AKT pathway, however, is inhibited with knockdown of METTL3. Our data indicate that ELA could act as a promoter of MSC proliferation and migration in vitro through the APJ receptor, something which might be attributed to the activation of the METTL3/PI3K/AKT signaling pathway. Therefore, ELA is a candidate for optimizing MSC-based cell therapy, while METTL3 is a potential target for its promoting action on MSCs.

Nonlinear regression of remaining surgery duration from videos via Bayesian LSTM-based deep negative correlation learning

In this paper, we address the problem of estimating remaining surgery duration (RSD) from surgical video frames. We propose a Bayesian long short-term memory (LSTM) network-based Deep Negative Correlation Learning approach called BD-Net for accurate regression of RSD prediction as well as estimation of prediction uncertainty. Our method aims to extract discriminative visual features from surgical video frames and model the temporal dependencies among frames to improve the RSD prediction accuracy. To this end, we propose to train an ensemble of Bayesian LSTMs on top of a backbone network by the way of deep negative correlation learning (DNCL). More specifically, we deeply learn a pool of decorrelated Bayesian regressors with sound generalization capabilities through managing their intrinsic diversities. BD-Net is simple and efficient. After training, it can produce both RSD prediction and uncertainty estimation in a single inference run. We demonstrate the efficacy of BD-Net on publicly available datasets of two different types of surgeries: one containing 101 cataract microscopic surgeries with short durations and the other containing 80 cholecystectomy laparoscopic surgeries with relatively longer durations. Experimental results on both datasets demonstrate that the proposed BD-Net achieves better results than the state-of-the-art (SOTA) methods. A reference implementation of our method can be found at: https://github.com/jywu511/BD-Net.

LAST: LAtent Space-Constrained Transformers for Automatic Surgical Phase Recognition and Tool Presence Detection

When developing context-aware systems, automatic surgical phase recognition and tool presence detection are two essential tasks. There exist previous attempts to develop methods for both tasks but majority of the existing methods utilize a frame-level loss function (e.g., cross-entropy) which does not fully leverage the underlying semantic structure of a surgery, leading to sub-optimal results. In this paper, we propose multi-task learning-based, LAtent Space-constrained Transformers, referred as LAST, for automatic surgical phase recognition and tool presence detection. Our design features a two-branch transformer architecture with a novel and generic way to leverage video-level semantic information during network training. This is done by learning a non-linear compact presentation of the underlying semantic structure information of surgical videos through a transformer variational autoencoder (VAE) and by encouraging models to follow the learned statistical distributions. In other words, LAST is of structure-aware and favors predictions that lie on the extracted low dimensional data manifold. Validated on two public datasets of the cholecystectomy surgery, i.e., the Cholec80 dataset and the M2cai16 dataset, our method achieves better results than other state-of-the-art methods. Specifically, on the Cholec80 dataset, our method achieves an average accuracy of 93.12±4.71%, an average precision of 89.25±5.49%, an average recall of 90.10±5.45% and an average Jaccard of 81.11 ±7.62% for phase recognition, and an average mAP of 95.15±3.87% for tool presence detection. Similar superior performance is also observed when LAST is applied to the M2cai16 dataset.

CholecTriplet2022: Show me a tool and tell me the triplet - An endoscopic vision challenge for surgical action triplet detection

Formalizing surgical activities as triplets of the used instruments, actions performed, and target anatomies is becoming a gold standard approach for surgical activity modeling. The benefit is that this formalization helps to obtain a more detailed understanding of tool-tissue interaction which can be used to develop better Artificial Intelligence assistance for image-guided surgery. Earlier efforts and the CholecTriplet challenge introduced in 2021 have put together techniques aimed at recognizing these triplets from surgical footage. Estimating also the spatial locations of the triplets would offer a more precise intraoperative context-aware decision support for computer-assisted intervention. This paper presents the CholecTriplet2022 challenge, which extends surgical action triplet modeling from recognition to detection. It includes weakly-supervised bounding box localization of every visible surgical instrument (or tool), as the key actors, and the modeling of each tool-activity in the form of ‹instrument, verb, target› triplet. The paper describes a baseline method and 10 new deep learning algorithms presented at the challenge to solve the task. It also provides thorough methodological comparisons of the methods, an in-depth analysis of the obtained results across multiple metrics, visual and procedural challenges; their significance, and useful insights for future research directions and applications in surgery.

EDRL: Entropy-guided disentangled representation learning for unsupervised domain adaptation in semantic segmentation

BACKGROUND AND OBJECTIVE

Deep learning-based approaches are excellent at learning from large amounts of data, but can be poor at generalizing the learned knowledge to testing datasets with domain shift, i.e., when there exists distribution discrepancy between the training dataset (source domain) and the testing dataset (target domain). In this paper, we investigate unsupervised domain adaptation (UDA) techniques to train a cross-domain segmentation method which is robust to domain shift, eliminating the requirement of any annotations on the target domain.

METHODS

To this end, we propose an Entropy-guided Disentangled Representation Learning, referred as EDRL, for UDA in semantic segmentation. Concretely, we synergistically integrate image alignment via disentangled representation learning with feature alignment via entropy-based adversarial learning into one network, which is trained end-to-end. We additionally introduce a dynamic feature selection mechanism via soft gating, which helps to further enhance the task-specific feature alignment. We validate the proposed method on two publicly available datasets: the CT-MR dataset and the multi-sequence cardiac MR (MS-CMR) dataset.

RESULTS

On both datasets, our method achieved better results than the state-of-the-art (SOTA) methods. Specifically, on the CT-MR dataset, our method achieved an average DSC of 84.8% when taking CT as the source domain and MR as the target domain, and an average DSC of 84.0% when taking MR as the source domain and CT as the target domain.

CONCLUSIONS

Results from comprehensive experiments demonstrate the efficacy of the proposed EDRL model for cross-domain medical image segmentation.

Pelvic tilt after Bernese periacetabular osteotomy-a long-term follow-up study

Patients with developmental dysplasia of the hip (DDH) are believed to present with increased anterior pelvic tilt to compensate for reduced anterior femoral head coverage. If true, pelvic tilt in dysplastic patients should be high preoperatively and decrease after correction with periacetabular osteotomy (PAO). To date, the evolution of pelvic tilt in long-term follow-up after PAO has not been reported. We therefore asked the following questions: (i) is there a difference in pelvic tilt between patients with DDH and an asymptomatic control group? (ii) How does pelvic tilt evolve during long-term follow-up after Bernese PAO compared with before surgery? This study is a therapeutic study with the level of evidence III. We retrospectively compared preoperative pelvic tilt in 64 dysplastic patients (71 hips) with an asymptomatic control group of 20 patients (20 hips). In addition, immediate postoperative and long-term follow-up (at 18 ± 8 [range 7-34 years) pelvic tilt was assessed and compared. Dysplastic patients had a significantly higher mean preoperative pelvic tilt than controls [2.3 ± 5.3° (-11.2° to 16.4°) versus 1.1 ± 3.0° (-4.9 to 5.9), P = 0.006]. Mean pelvic tilt postoperatively was 1.5 ± 5.3° (-11.2 to 17.0º, P = 0.221) and at long-term follow-up was 0.4 ± 5.7° (range -9.9° to 20.9°, P = 0.002). Dysplastic hips undergoing PAO show a statistically significant decrease in pelvic tilt during long-term follow-up. However, given the large interindividual variability in pelvic tilt, the observed differences may not achieve clinical significance.

CVCL: Context-aware Voxel-wise Contrastive Learning for label-efficient multi-organ segmentation

Despite the significant performance improvement on multi-organ segmentation with supervised deep learning-based methods, the label-hungry nature hinders their applications in practical disease diagnosis and treatment planning. Due to the challenges in obtaining expert-level accurate, densely annotated multi-organ datasets, label-efficient segmentation, such as partially supervised segmentation trained on partially labeled datasets or semi-supervised medical image segmentation, has attracted increasing attention recently. However, most of these methods suffer from the limitation that they neglect or underestimate the challenging unlabeled regions during model training. To this end, we propose a novel Context-aware Voxel-wise Contrastive Learning method, referred as CVCL, to take full advantage of both labeled and unlabeled information in label-scarce datasets for a performance improvement on multi-organ segmentation. Experimental results demonstrate that our proposed method achieves superior performance than other state-of-the-art methods.

[Efficacy of Hintermann calcaneal lengthening osteotomy for flexible flatfoot]

Objective: To investigate the clinical efficacy of Hintermann osteotomy (H-LCL) for flexible flatfoot. Methods: A follow-up study. Clinical data of 30 patients with flexible flatfoot treated with H-LCL operation from January 2020 to December 2021 in Sports Medical Center of the First Affiliated Hospital of Army Medical University were retrospectively analyzed. There were 8 males and 22 females, with a mean age of (39.0±15.2) years. The mean time from symptom onset to the diagnosis［M（Q₁，Q₃）］was 24.0 (5.5, 102.0) months. The functional and imaging scores of the patients before and after the last follow-up were compared to evaluate the clinical efficacy of the operation. The functional scores included American Orthopedic Foot and Ankle Society (AOFAS) score, visual analogue scale (VAS) of pain, pain interference (PI) and physical function (PF) index in Patient-Reported Outcomes Measurement Information System (PROMIS). And the imaging scores included Meary's angle, calcaneal pitch angle, calcaneal valgus angle and talonavicular coverage angle. Results: The mean operation time was (82.3±24.4) min, and the follow-up periods was (17.9±6.9) months. At the last follow-up, VAS of pain [M(Q₁, Q₃)] decreased from 5 (4, 6) to 2 (1, 2); PI decreased from 59.8±5.0 to 44.6±5.7; AOFAS increased from 65.2±10.0 to 85.8±3.3; PF increased from 50 (48.5,51.0) to 58.5 (54.0, 66.0); Meary's angle (antero-posterior image) decreased from 15.7° (10.1°, 29.2°) to 3.9° (2.6°, 5.3°); Meary's angle (lateral image) decreased from 13.5°±6.8° to 4.4°±2.6°; calcaneal pitch angle increased from 14.0°±3.3° to 18.6°±4.2°; calcaneal valgus angle decreased from 12.6°±7.3° to 4.3°±2.5°; and talonavicular coverage angle decreased from 20.9°±10.7° to 7.7°±5.2°. The up-mentioned parameters were all improved statistically significant at the last follow-up when compared with those before the operation (all P<0.05). Conclusion: H-LCL brings a significant improvement of clinical outcome scores and good radiological correction of flatfoot deformities in correcting flexible flatfoot, it conforms to the anatomical characteristics of the subtalar joint.

LatentPCN: latent space-constrained point cloud network for reconstruction of 3D patient-specific bone surface models from calibrated biplanar X-ray images

PURPOSE

Accurate three-dimensional (3D) models play crucial roles in computer assisted planning and interventions. MR or CT images are frequently used to derive 3D models but have the disadvantages that they are expensive or involving ionizing radiation (e.g., CT acquisition). An alternative method based on calibrated 2D biplanar X-ray images is highly desired.

METHODS

A point cloud network, referred as LatentPCN, is developed for reconstruction of 3D surface models from calibrated biplanar X-ray images. LatentPCN consists of three components: an encoder, a predictor, and a decoder. During training, a latent space is learned to represent shape features. After training, LatentPCN maps sparse silhouettes generated from 2D images to a latent representation, which is taken as the input to the decoder to derive a 3D bone surface model. Additionally, LatentPCN allows for estimation of a patient-specific reconstruction uncertainty.

RESULTS

We designed and conducted comprehensive experiments on datasets of 25 simulated cases and 10 cadaveric cases to evaluate the performance of LatentLCN. On these two datasets, the mean reconstruction errors achieved by LatentLCN were 0.83 mm and 0.92 mm, respectively. A correlation between large reconstruction errors and high uncertainty in the reconstruction results was observed.

CONCLUSION

LatentPCN can reconstruct patient-specific 3D surface models from calibrated 2D biplanar X-ray images with high accuracy and uncertainty estimation. The sub-millimeter reconstruction accuracy on cadaveric cases demonstrates its potential for surgical navigation applications.

[Correlation analysis and benchmark dose study on bone metabolic biochemical index of low doses of exposed hydrogen fluoride workers]

Objective: To analyze correlation of occupational hydrogen fluoride exposure to low doses of bone metabolism index through occupational epidemiological investigation and benchmark dose calculation. Methods: In May 2021, using cluster sampling method, 237 workers exposed to hydrogen fluoride in a company were selected as the contact group, and 83 workers not exposed to hydrogen fluoride in an electronics production company were selected as the control group. The external exposure dose and urinary fluoride concentration, blood and urine biochemical indicators of the workers was measured.The relationship between external dose and internal dose of hydrogen fluoride was analyzed. The external dose, urinary fluoride was used as exposure biomarkers, while serum osteocalcin (BGP), serum alkaline phosphatase (AKP) and urinary hydroxyproline (HYP) were used as effect biomarkers for bone metabolism of hydrogen fluoride exposure. The benchmark dose calculation software (BMDS1.3.2) was used to calculate benchmark dose (BMD) . Results: Urine fluoride concentration in the contact group was correlated with creatinine-adjusted urine fluoride concentration (r=0.69, P=0.001). There was no significant correlation between the external dose of hydrogen fluoride and urine fluoride in the contact group (r=0.03, P=0.132). The concentrations of urine fluoride in the contact group and the control group were (0.81±0.61) and (0.45±0.14) mg/L, respectively, and the difference between the two groups was statistically significant (t=5.01, P=0.025). Using BGP, AKP and HYP as effect indexes, the urinary BMDL-05 values were 1.28, 1.47 and 1.08 mg/L, respectively. Conclusion: Urinary fluoride can sensitively reflect the changes in the effect indexes of biochemical indexes of bone metabolism. BGP and HYP can be used as early sensitive effect indexes of occupational hydrogen fluoride exposure.

CholecTriplet2021: A benchmark challenge for surgical action triplet recognition

Context-aware decision support in the operating room can foster surgical safety and efficiency by leveraging real-time feedback from surgical workflow analysis. Most existing works recognize surgical activities at a coarse-grained level, such as phases, steps or events, leaving out fine-grained interaction details about the surgical activity; yet those are needed for more helpful AI assistance in the operating room. Recognizing surgical actions as triplets of ‹instrument, verb, target› combination delivers more comprehensive details about the activities taking place in surgical videos. This paper presents CholecTriplet2021: an endoscopic vision challenge organized at MICCAI 2021 for the recognition of surgical action triplets in laparoscopic videos. The challenge granted private access to the large-scale CholecT50 dataset, which is annotated with action triplet information. In this paper, we present the challenge setup and the assessment of the state-of-the-art deep learning methods proposed by the participants during the challenge. A total of 4 baseline methods from the challenge organizers and 19 new deep learning algorithms from the competing teams are presented to recognize surgical action triplets directly from surgical videos, achieving mean average precision (mAP) ranging from 4.2% to 38.1%. This study also analyzes the significance of the results obtained by the presented approaches, performs a thorough methodological comparison between them, in-depth result analysis, and proposes a novel ensemble method for enhanced recognition. Our analysis shows that surgical workflow analysis is not yet solved, and also highlights interesting directions for future research on fine-grained surgical activity recognition which is of utmost importance for the development of AI in surgery.

Measurement of pelvic tilt and rotation on AP radiographs using HipRecon: Validation and comparison to other parameters

In this paper, we present and evaluate HipRecon, a noncommercial software package that simultaneously calculates pelvic tilt and rotation from an anteroposterior pelvis radiograph. We asked: What is the (1) accuracy and precision, (2) robustness, and (3) intra-/interobserver reliability/reproducibility of HipRecon to analyze both pelvic tilt and rotation on conventional AP pelvis radiographs? (4) How does the prediction of pelvic tilt on AP pelvis radiographs using HipRecon compare to established measurement methods? We compared the actual pelvic tilt of 20 adult human cadaveric pelvises with the calculated pelvic orientation based on an AP pelvis radiograph using HipRecon software. The pelvises were mounted on a radiolucent fixture and a total of 380 AP pelvis radiographs with different configurations were acquired. In addition, we investigated the correlation between actual tilt and the tilt calculated using HipRecon and seven other established measurement methods. The calculated software accuracy was 0.2 ± 2.0° (-3.6-4.1) for pelvic tilt and 0.0 ± 1.2° (-2.2-2.3, p = 0.39) for pelvic rotation. The Bland-Altman analysis showed values that were evenly and randomly spread in both directions. HipRecon showed excellent consistency for the measurement of pelvic tilt and rotation (intraobserver intraclass-correlation coefficient [ICC]: 0.99 [95% CI: 0.99-0.99] and interobserver ICC 0.99 [95% CI: 0.99-0.99]). Of all eight analyzed methods, the highest correlation coefficient was found for HipRecon (r = 0.98, p < 0.001). In the future, HipRecon could be used to detect changes in patient-specific pelvic orientation, helping to improve clinical understanding and decision-making in pathologies of the hip.

The association between dietary selenium intake and Hashimoto's thyroiditis among US adults: National Health and Nutrition Examination Survey (NHANES), 2007-2012

BACKGROUND

Selenium has been shown to influence the pathological processes and physiological functions of thyroid. Although growing evidence has shown that selenium can improve the treatment of Hashimoto's thyroiditis (HT), there is a need to evaluate the association between dietary selenium intake and HT in a large cross-sectional study. This study explored the association between dietary selenium intake and HT based on the National Health reand Nutrition Examination Survey (NHANES) database (2007-2012).

METHODS

A total of 8756 of 30,442 participants were included in the study. Dietary selenium intake was the independent variable, while HT was the dependent variable. In addition, the relative importance of the selected variables was determined using the XGBoost model. A smooth curve was constructed based on the fully adjusted model to investigate the potential linear relationship between dietary selenium intake and HT. Smooth curves were also constructed to explore the linear/non-linear relationship between dietary selenium intake and thyroid peroxidase antibody (TPOAb)/ thyroglobulin antibody (TgAb).

RESULTS

The mean age of the enrolled participants was 44.35 years (± 20.92). The risk of HT was significantly reduced by a 35% per-unit increase in dietary selenium intake after fully adjusting for covariates according to the model (log2-transformed data; OR 0.65; 95% CI 0.51, 0.83). The XGBoost model revealed that dietary selenium intake was the most important variable associated with Hashimoto's thyroiditis. Dietary selenium intake (Log2-transformed) was negatively correlated with TPOAb levels [- 16.42 (- 22.18, - 10.65), P < 0.0001], while a non-linear relationship was observed between dietary selenium intake and TgAb with an inflection point of 6.58 (95.67 μg, Log2-transformed).

CONCLUSION

Dietary selenium intake is independently and inversely associated with HT risk. Moreover, dietary selenium intake is negatively correlated with TPOAb levels and non-linearly correlated with TGAb levels. Therefore, dietary selenium intake may be a safe and low-cost alternative for the prevention and treatment of HT.

Ultrasound Probe and Hand-Eye Calibrations for Robot-Assisted Needle Biopsy

In robot-assisted ultrasound-guided needle biopsy, it is essential to conduct calibration of the ultrasound probe and to perform hand-eye calibration of the robot in order to establish a link between intra-operatively acquired ultrasound images and robot-assisted needle insertion. Based on a high-precision optical tracking system, novel methods for ultrasound probe and robot hand-eye calibration are proposed. Specifically, we first fix optically trackable markers to the ultrasound probe and to the robot, respectively. We then design a five-wire phantom to calibrate the ultrasound probe. Finally, an effective method taking advantage of steady movement of the robot but without an additional calibration frame or the need to solve the AX=XB equation is proposed for hand-eye calibration. After calibrations, our system allows for in situ definition of target lesions and aiming trajectories from intra-operatively acquired ultrasound images in order to align the robot for precise needle biopsy. Comprehensive experiments were conducted to evaluate accuracy of different components of our system as well as the overall system accuracy. Experiment results demonstrated the efficacy of the proposed methods.

A Novel Point Set Registration-Based Hand-Eye Calibration Method for Robot-Assisted Surgery

Pedicle screw insertion with robot assistance dramatically improves surgical accuracy and safety when compared with manual implantation. In developing such a system, hand-eye calibration is an essential component that aims to determine the transformation between a position tracking and robot-arm systems. In this paper, we propose an effective hand-eye calibration method, namely registration-based hand-eye calibration (RHC), which estimates the calibration transformation via point set registration without the need to solve the AX=XB equation. Our hand-eye calibration method consists of tool-tip pivot calibrations in two-coordinate systems, in addition to paired-point matching, where the point pairs are generated via the steady movement of the robot arm in space. After calibration, our system allows for robot-assisted, image-guided pedicle screw insertion. Comprehensive experiments are conducted to verify the efficacy of the proposed hand-eye calibration method. A mean distance deviation of 0.70 mm and a mean angular deviation of 0.68° are achieved by our system when the proposed hand-eye calibration method is used. Further experiments on drilling trajectories are conducted on plastic vertebrae as well as pig vertebrae. A mean distance deviation of 1.01 mm and a mean angular deviation of 1.11° are observed when the drilled trajectories are compared with the planned trajectories on the pig vertebrae.

[A multicenter epidemiological study of acute bacterial meningitis in children]

Objective: To analyze the clinical epidemiological characteristics including composition of pathogens , clinical characteristics, and disease prognosis acute bacterial meningitis (ABM) in Chinese children. Methods: A retrospective analysis was performed on the clinical and laboratory data of 1 610 children <15 years of age with ABM in 33 tertiary hospitals in China from January 2019 to December 2020. Patients were divided into different groups according to age,<28 days group, 28 days to <3 months group, 3 months to <1 year group, 1-<5 years of age group, 5-<15 years of age group; etiology confirmed group and clinically diagnosed group according to etiology diagnosis. Non-numeric variables were analyzed with the Chi-square test or Fisher's exact test, while non-normal distrituction numeric variables were compared with nonparametric test. Results: Among 1 610 children with ABM, 955 were male and 650 were female (5 cases were not provided with gender information), and the age of onset was 1.5 (0.5, 5.5) months. There were 588 cases age from <28 days, 462 cases age from 28 days to <3 months, 302 cases age from 3 months to <1 year of age group, 156 cases in the 1-<5 years of age and 101 cases in the 5-<15 years of age. The detection rates were 38.8% (95/245) and 31.5% (70/222) of Escherichia coli and 27.8% (68/245) and 35.1% (78/222) of Streptococcus agalactiae in infants younger than 28 days of age and 28 days to 3 months of age; the detection rates of Streptococcus pneumonia, Escherichia coli, and Streptococcus agalactiae were 34.3% (61/178), 14.0% (25/178) and 13.5% (24/178) in the 3 months of age to <1 year of age group; the dominant pathogens were Streptococcus pneumoniae and the detection rate were 67.9% (74/109) and 44.4% (16/36) in the 1-<5 years of age and 5-<15 years of age . There were 9.7% (19/195) strains of Escherichia coli producing ultra-broad-spectrum β-lactamases. The positive rates of cerebrospinal fluid (CSF) culture and blood culture were 32.2% (515/1 598) and 25.0% (400/1 598), while 38.2% (126/330)and 25.3% (21/83) in CSF metagenomics next generation sequencing and Streptococcus pneumoniae antigen detection. There were 4.3% (32/790) cases of which CSF white blood cell counts were normal in etiology confirmed group. Among 1 610 children with ABM, main intracranial imaging complications were subdural effusion and (or) empyema in 349 cases (21.7%), hydrocephalus in 233 cases (14.5%), brain abscess in 178 cases (11.1%), and other cerebrovascular diseases, including encephalomalacia, cerebral infarction, and encephalatrophy, in 174 cases (10.8%). Among the 166 cases (10.3%) with unfavorable outcome, 32 cases (2.0%) died among whom 24 cases died before 1 year of age, and 37 cases (2.3%) had recurrence among whom 25 cases had recurrence within 3 weeks. The incidences of subdural effusion and (or) empyema, brain abscess and ependymitis in the etiology confirmed group were significantly higher than those in the clinically diagnosed group (26.2% (207/790) vs. 17.3% (142/820), 13.0% (103/790) vs. 9.1% (75/820), 4.6% (36/790) vs. 2.7% (22/820), χ²=18.71, 6.20, 4.07, all P<0.05), but there was no significant difference in the unfavorable outcomes, mortility, and recurrence between these 2 groups (all P>0.05). Conclusions: The onset age of ABM in children is usually within 1 year of age, especially <3 months. The common pathogens in infants <3 months of age are Escherichia coli and Streptococcus agalactiae, and the dominant pathogen in infant ≥3 months is Streptococcus pneumoniae. Subdural effusion and (or) empyema and hydrocephalus are common complications. ABM should not be excluded even if CSF white blood cell counts is within normal range. Standardized bacteriological examination should be paid more attention to increase the pathogenic detection rate. Non-culture CSF detection methods may facilitate the pathogenic diagnosis.

What Factors Are Associated With Postoperative Ischiofemoral Impingement After Bernese Periacetabular Osteotomy in Developmental Dysplasia of the Hip?

BACKGROUND

Any abnormal structures that contribute to the narrowing of the ischiofemoral space could induce ischiofemoral impingement. Bernese periacetabular osteotomy (PAO) medializes the hip center and, therefore, decreases contact stress on the cartilage in developmental dysplasia of the hip (DDH). However, medialization of the hip center might also narrow the ischiofemoral space, which may increase the risk of postoperative ischiofemoral impingement in patients with acetabular dysplasia who are undergoing PAO. Furthermore, the dysplastic hip has less ischiofemoral space and less space for the quadratus femoris. A few studies have focused on the amount of medialization of the hip center, but the proportion of postoperative ischiofemoral impingement after PAO has not been investigated.

QUESTIONS/PURPOSES

(1) What proportion of patients develop ischiofemoral impingement after undergoing unilateral PAO for DDH? (2) What radiographic factors are associated with postoperative ischiofemoral impingement in patients who underwent PAO for DDH? (3) How much hip center medialization is safe so as to avoid postoperative ischiofemoral impingement during PAO?

METHODS

Between 2014 and 2016, we treated 265 adult patients who had symptomatic residual acetabular dysplasia (lateral center-edge angle less than 20°) using PAO. During that time, we generally offered PAO to patients with acetabular dysplasia when the patients had no advanced osteoarthritis (Tönnis grade < 2). Of those, we considered only patients who underwent primary PAO without femoral osteotomy as potentially eligible. Based on that, 65% (173 of 265) were eligible; a further 9% (24 of 265) were excluded due to leg length discrepancy, spine disorders, or joint replacement in the contralateral side, and another 6% (17 of 265) of patients were lost before the minimum study follow-up of 2 years or had incomplete datasets, leaving 50% (132 of 265) for analysis in this retrospective study at a mean of 2.70 ± 0.71 years. The diagnosis of ischiofemoral impingement was defined by symptoms, MRI, and diagnostic ischiofemoral injection. We ascertained the percentage of patients with this diagnosis to answer the first research question. To answer the second question, we divided the patients into two groups: PAO patients with ischiofemoral impingement and PAO patients without ischiofemoral impingement. The demographic data and preoperative imaging parameters of patients in both groups were compared. There were statistical differences in acetabular version, ischial angle, neck-shaft angle, the presence of positive coxa profunda sign, McKibbin index, ischiofemoral space, quadratus femoris space, anterior acetabular section angle, and the net amount of hip center medialization. To investigate potential factors associated with postoperative ischiofemoral impingement in patients who underwent PAO, these factors underwent binary logistic regression analysis. To answer the third question, the cutoff value of the net amount of hip center medialization was evaluated using receiver operator characteristic curve and the Youden index method.

RESULTS

We found that 26% (35 of 132) of PAO dysplastic hips had postoperative ischiofemoral impingement. After controlling for confounding variables such as acetabular version, ischial angle, femoral neck version, McKibbin index, and ischiofemoral space, we found that an increasing neck-shaft angle (odds ratio 1.14 [95% confidence interval 1.01 to 1.29]; p = 0.03), a positive coxa profunda sign (OR 0.13 [95% CI 0.03 to 0.58]; p < 0.01), and an increasing net amount of hip center medialization (OR 2.76 [95% CI 1.70 to 4.47]; p < 0.01) were associated with postoperative ischiofemoral impingement in patients with DDH who underwent PAO (R 2 = 0.73). The cutoff values of neck-shaft angle was 138.4°. The cutoff values of the net amount of hip center medialization was 1.9 mm.

CONCLUSIONS

Postoperative ischiofemoral impingement could occur in patients with acetabular dysplasia who have undergone PAO after hip center medialization. An increasing neck-shaft angle, a positive coxa profunda sign on preoperative imaging, and excessive medialization of the hip center are factors associated with ischiofemoral impingement development in these patients. Therefore, we suggest that physicians measure the ischiofemoral space on a preoperative CT when patients with DDH have an increasing neck-shaft angle (> 138.4°) or a positive coxa profunda sign on radiological imaging. During PAO, the amount of hip center medialization should be carefully controlled to keep these patients from developing postoperative ischiofemoral impingement.

LEVEL OF EVIDENCE

Level III, therapeutic study.

[Analysis of the effectiveness of coracoid osteotomy and concentric coaxial reconstruction of the glenoid cavity in the treatment of recurrent anterior dislocation of the shoulder joint]

Objective: To investigate the clinical efficacy of the modified Latarjet procedure in the treatment of recurrent anterior subluxation of the shoulder by "coaxial co-arc" reconstruction of the glenoid cavity. Methods: The clinical data of 103 cases (106 shoulders) of recurrent anterior dislocation of the shoulder admitted to the First Affiliated Hospital of the Army Military Medical University from January 2005 to December 2020 were retrospectively studied. Out of these cases, 84 were males and 19 were females; 31 with left-sided injuries while 75 with right-sided injuries, with a mean age of (29.4±11.5) years (16-61 years). The preoperative anterior fear test was positive, and a modified Latarjet procedure was used to reconstruct the shoulder glenoid defect through a "coaxial co-arc". The Rowe score, simple shoulder test (SST) score, and Visual analogue scale (VAS) score of pain were used to assess the shoulder's function. Parameters such as the postoperative shoulder recurrent dislocation rate, shoulder body external rotation angle, and subscapularis muscle strength changes were recorded postoperatively. Moreover, radiographs and CT scans were used to check for the incidence of osteoarthritis (Samson-Prieto score). Results: After a mean follow-up of 9.0 years (1 to 16 years), bony healing occurred 3 to 6 months postoperatively. The Rowe score improved from 40.4±6.5 preoperatively to 93.2±2.5 (P<0.001), the SST score improved from 5.2±1.3 preoperatively to 10.1±1.5 (P<0.001), and the VAS pain score decreased from 3.5±1.9 preoperatively to 1.1±1.2 (P<0.001) at the final follow-up. The angle of lateral external rotation of the shoulder joint was 58.8°±15.6° preoperatively and 57.6°±14.5° postoperatively with no statistically significant difference (P>0.05). There was no statistically significant difference in the measurement of subscapularis muscle strength between the healthy side and the affected side (P>0.05). In 89.6% of patients after surgery, coaxial co-arc reconstruction of the shoulder glenoid was obtained, and the shoulder glenoid defect and postoperative inclusion angle were significantly improved compared with those before surgery (P<0.001). Postoperatively, new-onset osteoarthritis developed in 7 cases (7/98), arthritis progressed in 2 cases (2/8), incisional healing was poor in 2 cases (2/98), and revision surgery was performed in 2 cases (2/98) due to bone mass detachment. Conclusion: Coracoid osteotomy and concentric coaxial reconstruction of the glenoid cavity elicits adequate good clinical efficacy for cases of recurrent anterior shoulder dislocation, with low recurrence rates, low revision rates and low incidence of osteoarthritis.

Translocator protein (18 kDa) regulates the microglial phenotype in Parkinson's disease through P47

Numerous studies have suggested that the phenotypic transformation of microglia plays a role in the pathogenesis of Parkinson's disease (PD). Translocator protein (TSPO) is an 18 kDa translocator membrane protein that acts as a marker of neuroinflammation and suppresses neuroinflammation; however, its underlying mechanism remains unclear. Although TSPO ligands were found to be protective in several neurodegenerative paradigms, few studies have evaluated their effects on microglial polarization, and underlying mechanisms need to be explored. In the present study, we examined the effects of TSPO and PK11195, a TSPO ligand, on lipopolysaccharide (LPS)+interferon (IFN)-γ-induced inflammatory factors and oxidative stress in microglia using enzyme-linked immunosorbent assay. The effect of TSPO and PK11195 on LPS+IFN-γ-induced microglial cell apoptosis was examined using immunofluorescence (IF), flow cytometry, and western blotting. The interaction between TSPO and P47 was investigated using IF and co-immunoprecipitation analysis. In vivo experiments confirmed the influence of TSPO and its ligand on motility, a-Syn, and dopaminergic neuronal damage. Our findings indicate that TSPO may regulate the microglial phenotype in PD via P47, suggesting a potential role in anti-PD therapy.

Handling Imbalanced Data: Uncertainty-guided Virtual Adversarial Training with Batch Nuclear-norm Optimization for Semi-supervised Medical Image Classification

In many clinical settings, a lot of medical image datasets suffer from imbalance problems, which makes predictions of trained models to be biased toward majority classes. Semi-supervised Learning (SSL) algorithms trained with such imbalanced datasets become more problematic since pseudo-supervision of unlabeled data are generated from the model's biased predictions. To address these issues, in this work, we propose a novel semi-supervised deep learning method, i.e., uncertainty-guided virtual adversarial training (VAT) with batch nuclear-norm (BNN) optimization, for large-scale medical image classification. To effectively exploit useful information from both labeled and unlabeled data, we leverage VAT and BNN optimization to harness the underlying knowledge, which helps to improve discriminability, diversity and generalization of the trained models. More concretely, our network is trained by minimizing a combination of four types of losses, including a supervised cross-entropy loss, a BNN loss defined on the output matrix of labeled data batch (lBNN loss), a negative BNN loss defined on the output matrix of unlabeled data batch (uBNN loss), and a VAT loss on both labeled and unlabeled data. We additionally propose to use uncertainty estimation to filter out unlabeled samples near the decision boundary when computing the VAT loss. We conduct comprehensive experiments to evaluate the performance of our method on two publicly available datasets and one in-house collected dataset. The experimental results demonstrated that our method achieved better results than state-of-the-art SSL methods.

Deep learning-based 2D/3D registration of an atlas to biplanar X-ray images

Purpose

The registration of a 3D atlas image to 2D radiographs enables 3D pre-operative planning without the need to acquire costly and high-dose CT-scans. Recently, many deep-learning-based 2D/3D registration methods have been proposed which tackle the problem as a reconstruction by regressing the 3D image immediately from the radiographs, rather than registering an atlas image. Consequently, they are less constrained against unfeasible reconstructions and have no possibility to warp auxiliary data. Finally, they are, by construction, limited to orthogonal projections.

Methods

We propose a novel end-to-end trainable 2D/3D registration network that regresses a dense deformation field that warps an atlas image such that the forward projection of the warped atlas matches the input 2D radiographs. We effectively take the projection matrix into account in the regression problem by integrating a projective and inverse projective spatial transform layer into the network.

Results

Comprehensive experiments conducted on simulated DRRs from patient CT images demonstrate the efficacy of the network. Our network yields an average Dice score of 0.94 and an average symmetric surface distance of 0.84 mm on our test dataset. It has experimentally been determined that projection geometries with 80\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$^{\circ }$$\end{document}∘ to 100\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$^{\circ }$$\end{document}∘ projection angle difference result in the highest accuracy.

Conclusion

Our network is able to accurately reconstruct patient-specific CT-images from a pair of near-orthogonal calibrated radiographs by regressing a deformation field that warps an atlas image or any other auxiliary data. Our method is not constrained to orthogonal projections, increasing its applicability in medical practices. It remains a future task to extend the network for uncalibrated radiographs.

MCG-Net: End-to-end Fine-grained Delineation and Diagnostic Classification of Cardiac Events from Magnetocardiographs

In this paper, we propose an end-to-end deep learning architecture, referred as MCG-Net, integrating convolutional neural network (CNN) with transformer-based global context block for fine-grained delineation and diagnostic classification of four cardiac events from magnetocardiogram (MCG) data, namely Q-, R-, S- and T-waves. MCG-Net} takes advantage of a multi-resolution CNN backbone as well as the state-of-the-art (SOTA) transformer encoders that facilitate global temporal feature aggregation. Besides the novel network architecture, we introduce a multi-task learning scheme to achieve simultaneous delineation and classification. Specifically, the problem of MCG delineation is formulated as multi-class heatmap regression. Meanwhile, a binary diagnostic classification label as well as a duration are jointly estimated for each cardiac event using features that are temporally aligned by event heatmaps. The framework is evaluated on a clinical MCG dataset, containing data collected from 270 subjects with cardiac anomalies and 108 control subjects. We designed and conducted a two-fold cross-validation study to validate the proposed method and to compare its performance with the SOTA methods. Experimental results demonstrated that our method outperformed counterparts on both event delineation and diagnostic classification tasks, achieving respectively an average ECG-F1 of 0.987 and an average Event-F1 of 0.975 for MCG delineation, and an average accuracy of 0.870, an average sensitivity of 0.732, an average specificity of 0.914 and an average AUC of 0.903 for diagnostic classification. Comprehensive ablation experiments are additionally performed to investigate effectiveness of different network components.

Spine-transformers: Vertebra labeling and segmentation in arbitrary field-of-view spine CTs via 3D transformers

In this paper, we address the problem of fully automatic labeling and segmentation of 3D vertebrae in arbitrary Field-Of-View (FOV) CT images. We propose a deep learning-based two-stage solution to tackle these two problems. More specifically, in the first stage, the challenging vertebra labeling problem is solved via a novel transformers-based 3D object detector that views automatic detection of vertebrae in arbitrary FOV CT scans as a one-to-one set prediction problem. The main components of the new method, called Spine-Transformers, are a one-to-one set based global loss that forces unique predictions and a light-weighted 3D transformer architecture equipped with a skip connection and learnable positional embeddings for encoder and decoder, respectively. We additionally propose an inscribed sphere-based object detector to replace the regular box-based object detector for a better handling of volume orientation variations. Our method reasons about the relationships of different levels of vertebrae and the global volume context to directly infer all vertebrae in parallel. In the second stage, the segmentation of the identified vertebrae and the refinement of the detected centers are then done by training one single multi-task encoder-decoder network for all vertebrae as the network does not need to identify which vertebra it is working on. The two tasks share a common encoder path but with different decoder paths. Comprehensive experiments are conducted on two public datasets and one in-house dataset. The experimental results demonstrate the efficacy of the present approach.

MiR-194-5p enhances the sensitivity of nonsmall-cell lung cancer to doxorubicin through targeted inhibition of hypoxia-inducible factor-1

Background

Despite chemotherapy being a common treatment, an increase in chemoresistance over time is unavoidable. We therefore investigated the role of miR-194-5p in regulating chordoma cell behavior and examined the downstream effectors of miR-194-5p.

Methods

In this study, NSCLC cell lines A549 and H460 were cultured under hypoxic conditions for 1 week to induce drug resistance to doxorubicin (DOX). The connection between miR-194-5p and HIF-1 was revealed by reverse transcription and real-time polymerase chain reaction (RT-qPCR), western blot, and dual-luciferase assays. We used TUNEL staining and the CCK-8 test to assess the sensitivity of NSCLC cells to DOX.

Results

We found that hypoxia-induced NSCLC cells enhanced resistance to DOX. MiR-194-5p was substantially reduced, and HIF-1 was increased in hypoxia-induced drug-resistant NSCLC cells. Moreover, miR-194-5p successfully induced NSCLC cell apoptosis by directly inhibiting HIF-1, thereby enhancing DOX sensitivity.

Conclusions

MiR-194-5p enhanced the sensitivity of NSCLC cells to DOX by directly inhibiting HIF-1. This work provides insights into underlying treatments for drug-resistant NSCLC.

Does the Rule of Thirds Adequately Detect Deficient and Excessive Acetabular Coverage?

BACKGROUND

Assessment of AP acetabular coverage is crucial for choosing the right surgery indication and for obtaining a good outcome after hip-preserving surgery. The quantification of anterior and posterior coverage is challenging and requires either other conventional projections, CT, MRI, or special measurement software, which is cumbersome, not widely available and implies additional radiation. We introduce the "rule of thirds" as a promising alternative to provide a more applicable and easy method to detect an excessive or deficient AP coverage. This method attributes the intersection point of the anterior (posterior) wall to thirds of the femoral head radius (diameter), the medial third suggesting deficient and the lateral third excessive coverage.

QUESTION/PURPOSE

What is the validity (area under the curve [AUC], sensitivity, specificity, positive/negative likelihood ratios [LR(+)/LR(-)], positive/negative predictive values [PPV, NPV]) for the rule of thirds to detect (1) excessive and (2) deficient anterior and posterior coverages compared with previously established radiographic values of under-/overcoverage using Hip2Norm as the gold standard?

METHODS

We retrospectively evaluated all consecutive patients between 2003 and 2015 from our institutional database who were referred to our hospital for hip pain and were potentially eligible for joint-preserving hip surgery. We divided the study group into six specific subgroups based on the respective acetabular pathomorphology to cover the entire range of anterior and posterior femoral coverage (dysplasia, overcoverage, severe overcoverage, excessive acetabular anteversion, acetabular retroversion, total acetabular retroversion). From this patient cohort, 161 hips were randomly selected for analysis. Anterior and posterior coverage was determined with Hip2Norm, a validated computer software program for evaluating acetabular morphology. The anterior and posterior wall indices were measured on standardized AP pelvis radiographs, and the rule of thirds was applied by one observer.

RESULTS

The detection of excessive anterior and posterior acetabular wall using the rule of thirds revealed an AUC of 0.945 and 0.933, respectively. Also the detection of a deficient anterior and posterior acetabular wall by applying the rule of thirds revealed an AUC of 0.962 and 0.876, respectively. For both excessive and deficient anterior and posterior acetabular coverage, we found high specificities and PPVs but low sensitivities and NPVs.

CONCLUSION

We found a high probability for an excessive (deficient) acetabular wall when this intersection point lies in the lateral (medial) third, which would qualify for surgical correction. On the other hand, if this point is not in the lateral (medial) third, an excessive (deficient) acetabular wall cannot be categorically excluded. Thus, the rule of thirds is very specific but not as sensitive as we had expected.

LEVEL OF EVIDENCE

Level II, diagnostic study.

A Projector-Based Augmented Reality Navigation System for Computer-Assisted Surgery

In the medical field, guidance to follow the surgical plan is crucial. Image overlay projection is a solution to link the surgical plan with the patient. It realizes augmented reality (AR) by projecting computer-generated image on the surface of the target through a projector, which can visualize additional information to the scene. By overlaying anatomical information or surgical plans on the surgery area, projection helps to enhance the surgeon's understanding of the anatomical structure, and intuitively visualizes the surgical target and key structures of the operation, and avoid the surgeon's sight diversion between monitor and patient. However, it still remains a challenge to project the surgical navigation information on the target precisely and efficiently. In this study, we propose a projector-based surgical navigation system. Through the gray code-based calibration method, the projector can be calibrated with a camera and then be integrated with an optical spatial locator, so that the navigation information of the operation can be accurately projected onto the target area. We validated the projection accuracy of the system through back projection, with average projection error of 3.37 pixels in x direction and 1.51 pixels in y direction, and model projection with an average position error of 1.03 ± 0.43 mm, and carried out puncture experiments using the system with correct rate of 99%, and qualitatively analyzed the system's performance through the questionnaire. The results demonstrate the efficacy of our proposed AR system.

[LIM-domain binding protein 2 regulated by m6A modification inhibits lung adenocarcinoma cell proliferation in vitro]

OBJECTIVE

To investigate the role and expression pattern of LIM-domain binding protein 2 (LDB2) in lung adenocarcinoma.

OBJECTIVE

We studied the expression pattern of LDB2 in lung adenocarcinoma based on data from the online databases TCGA, GEO and CPTAC, and the results were verified in lung adenocarcinoma tissues and cells using immunohistochemistry, qRT-PCR and Western blotting. The relationship between LDB2 and the prognosis of patients with lung adenocarcinoma was analyzed using GEPIA and GEO databases. We further analyzed the role of LDB2 in regulating cell behaviors in a H1299 cell model over-expressing LDB2 using cell counting, soft agar colony forming assay and flow cytometry. The m⁶A binding sites on LDB2 were confirmed by bioinformatics analysis and MeRIP-qPCR assays. The effect of YTHDC2 on LDB2 was examined using qRT-PCR and Western blotting, and the binding of YTHDC2 to the transcript of LDB2 was verified with RIP-qPCR assays. Dual luciferase reporter assay was performed to verify YTHDC2 functioning via m⁶A sites.

OBJECTIVE

LDB2 expression was significantly decreased in lung adenocarcinoma in comparison with normal tissues based on data from TCGA, GEPIA and CPTAC, and the same results were obtained from 80 lung adenocarcinoma tissues and 17 adjacent normal tissues. Similarly, LDB2 expression was decreased in lung adenocarcinoma cells as compared with 16HBE cells. The data from Prognoscan and GEPIA suggested that a high LDB2 expression was positively correlated with a more favorable outcome of lung adenocarcinoma patients. LDB2-overexpressing H1299 cells showed a significant inhibition of proliferative activity with cell cycle arrest in S phage. Bioinformatics analysis and MeRIP-qPCR assay confirmed the presence of m⁶A sites on LDB2. The m⁶A reader YTHDC2 was positively related with LDB2 in lung adenocarcinoma based on data from GEPIA (r=0.22). Overexpression YTHDC2 significantly enhanced LDB2 expression in H1299 cells by about 19.35 folds. Dual luciferase reporter assay showed that YTHDC2 enhanced the promoter activity in the wild-type group but not in deletion group.

OBJECTIVE

LDB2 expression can be up-regulated by m⁶A reader YTHDC2 in lung adenocarcinoma to inhibit the proliferation of the tumor cells in vitro.

MiR-24 inhibits oligodendrocyte precursor cell differentiation after spinal injury by targeting adrenal medulla

OBJECTIVE

Oligodendrocyte precursor cells (OPCs) differentiate into oligodendrocytes (OLs) that provide nutrients to neurons. Adrenal medulla is (ADM) involved in nerve damage. MiR-24 participates in various diseases. However, the regulation and mechanism of miR-24 in oligodendrocyte precursor cell differentiation after spinal injury is unclear. MATERIALS  AND METHODS: Wistar rats were divided into sham operation group and model group. Real Time-PCR detects miR-24, PDGFRa and NG2 and MBP expression. OPC cells were cultured and divided into control group, miR-24 group, and si-miR-24 group followed by analysis of miR-24 expression by Real Time-PCR, expression of PDGFRa, NG2 and MBP by Western blot, as well as ADM content and secretion of IL-6 and TNF-α by enzyme-linked immunosorbent assay (ELISA).

RESULTS

Expression of miR-24, PDGFRa, and NG2 was increased in the model group and MBP and ADM expression was decreased with increased secretion of IL-6 and TNF-α. Compared with control group, the difference was statistically significant (p<0.05). Upregulation of miR-24 promoted the expression of PDGFRa and NG2, decreased MBP and ADM level, and increased IL-6 and TNF-α secretion. Compared with control group, the difference was statistically significant (p<0.05). Downregulation of miR-24 reversed the above changes, and the difference was statistically significant (p<0.05).

CONCLUSIONS

MiR-24 expression is increased in spinal injury. Upregulation of miR-24 expression reduces adrenal medulla expression and inhibits oligodendrocyte precursor cell differentiation.

Chemopreventive effect of galangin against benzo(a)pyrene-induced stomach tumorigenesis through modulating aryl hydrocarbon receptor in Swiss albino mice

The present study was aimed to evaluate the chemopreventive potential of galangin against benzo(a)pyrene (BaP)-induced stomach carcinogenesis in Swiss albino mice. Stomach cancer was induced in experimental mice using BaP oral administration. The mice were treated with galangin (10 mg/kg b.wt.) before and during BaP administration. Oral administration of galangin at a dose of 10 mg/kg b.wt. significantly (p < 0.05) prevented the tumor incidence, tumor volume in the experimental animals. Further, galangin pretreatment prevents BaP-induced lipid peroxidation and restores BaP-mediated loss of cellular antioxidants status. It has also been found that galangin prevents BaP-induced activation of phase I detoxification enzymes. Furthermore, galangin pretreatment prevented the BaP-induced overexpression of cytochrome P450s isoform genes (CYP1A1, CYP1B1), aryl hydrocarbon receptor system (AhR, ARNT), transcriptional activators (CBP/p300, NF-kB), tumor growth factors, proto-oncogenes, invasion markers (TGFB, SRC-1, MYC, iNOS, MMP2, MMP9) and Phase II metabolic isoenzyme genes (GST) in the stomach tissue homogenate when compared to the control groups. The western blot results confirm that galangin (10 mg/kg. b.wt.) treatment significantly prevented the BaP-mediated expression of ArR, ARNT, and CYP1A1 proteins in the mouse stomach tissue. Therefore, the present results confirm that galangin prevents BaP-induced stomach carcinogenesis probably through modulating ArR and ARNT expression in the experimental mice.

EquiSim: An Open-Source Articulatable Statistical Model of the Equine Distal Limb

Most digital models of the equine distal limb that are available in the community are static and/or subject specific; hence, they have limited applications in veterinary research. In this paper, we present an articulatable model of the entire equine distal limb based on statistical shape modeling. The model describes the inter-subject variability in bone geometry while maintaining proper jointspace distances to support model articulation toward different poses. Shape variation modes are explained in terms of common biometrics in order to ease model interpretation from a veterinary point of view. The model is publicly available through a graphical user interface (https://github.com/jvhoutte/equisim) in order to facilitate future digitalization in veterinary research, such as computer-aided designs, three-dimensional printing of bone implants, bone fracture risk assessment through finite element methods, and data registration and segmentation problems for clinical practices.

Nonlinear Regression via Deep Negative Correlation Learning

Nonlinear regression has been extensively employed in many computer vision problems (e.g., crowd counting, age estimation, affective computing). Under the umbrella of deep learning, two common solutions exist i) transforming nonlinear regression to a robust loss function which is jointly optimizable with the deep convolutional network, and ii) utilizing ensemble of deep networks. Although some improved performance is achieved, the former may be lacking due to the intrinsic limitation of choosing a single hypothesis and the latter may suffer from much larger computational complexity. To cope with those issues, we propose to regress via an efficient "divide and conquer" manner. The core of our approach is the generalization of negative correlation learning that has been shown, both theoretically and empirically, to work well for non-deep regression problems. Without extra parameters, the proposed method controls the bias-variance-covariance trade-off systematically and usually yields a deep regression ensemble where each base model is both "accurate" and "diversified." Moreover, we show that each sub-problem in the proposed method has less Rademacher Complexity and thus is easier to optimize. Extensive experiments on several diverse and challenging tasks including crowd counting, personality analysis, age estimation, and image super-resolution demonstrate the superiority over challenging baselines as well as the versatility of the proposed method. The source code and trained models are available on our project page: https://mmcheng.net/dncl/.

MRI-based 3D models of the hip joint enables radiation-free computer-assisted planning of periacetabular osteotomy for treatment of hip dysplasia using deep learning for automatic segmentation

Introduction

Both Hip Dysplasia(DDH) and Femoro-acetabular-Impingement(FAI) are complex three-dimensional hip pathologies causing hip pain and osteoarthritis in young patients. 3D-MRI-based models were used for radiation-free computer-assisted surgical planning. Automatic segmentation of MRI-based 3D-models are preferred because manual segmentation is time-consuming.To investigate(1) the difference and(2) the correlation for femoral head coverage(FHC) between automatic MR-based and manual CT-based 3D-models and (3) feasibility of preoperative planning in symptomatic patients with hip diseases.

Methods

We performed an IRB-approved comparative, retrospective study of 31 hips(26 symptomatic patients with hip dysplasia or FAI). 3D MRI sequences and CT scans of the hip were acquired. Preoperative MRI included axial-oblique T1 VIBE sequence(0.8 mm³ isovoxel) of the hip joint. Manual segmentation of MRI and CT scans were performed. Automatic segmentation of MRI-based 3D-models was performed using deep learning.

Results

(1)The difference between automatic and manual segmentation of MRI-based 3D hip joint models was below 1 mm(proximal femur 0.2 ± 0.1 mm and acetabulum 0.3 ± 0.5 mm). Dice coefficients of the proximal femur and the acetabulum were 98 % and 97 %, respectively. (2)The correlation for total FHC was excellent and significant(r = 0.975, p < 0.001) between automatic MRI-based and manual CT-based 3D-models. Correlation for total FHC (r = 0.979, p < 0.001) between automatic and manual MR-based 3D models was excellent.(3)Preoperative planning and simulation of periacetabular osteotomy was feasible in all patients(100 %) with hip dysplasia or acetabular retroversion.

Conclusions

Automatic segmentation of MRI-based 3D-models using deep learning is as accurate as CT-based 3D-models for patients with hip diseases of childbearing age. This allows radiation-free and patient-specific preoperative simulation and surgical planning of periacetabular osteotomy for patients with DDH.

Evaluation of CT-MR image registration methodologies for 3D preoperative planning of forearm surgeries

Computerized surgical planning for forearm procedures that considers both soft and bony tissue, requires alignment of preoperatively acquired computed tomography (CT) and magnetic resonance (MR) images by image registration. Normalized mutual information (NMI) registration techniques have been researched to improve efficiency and to eliminate the user dependency associated with manual alignment. While successfully applied in various medical fields, the application of NMI registration to images of the forearm, for which the relative pose of the radius and ulna likely differs between CT and MR acquisitions, is yet to be described. To enable the alignment of CT and MR forearm data, we propose an NMI-based registration pipeline, which allows manual steering of the registration algorithm to the desired image subregion and is, thus, applicable to the forearm. Successive automated registration is proposed to enable planning incorporating multiple target anatomical structures such as the radius and ulna. With respect to gold-standard manual registration, the proposed registration methodology achieved mean accuracies of 0.08 ± 0.09 mm (0.01-0.41 mm range) in comparison with 0.28 ± 0.23 mm (0.03-0.99 mm range) associated with a landmark-based registration when tested on 40 patient data sets. Application of the proposed registration pipeline required less than 10 minutes on average compared with 20 minutes required by the landmark-based registration. The clinical feasibility and relevance of the method were tested on two different clinical applications, a forearm tumor resection and radioulnar joint instability analysis, obtaining accurate and robust CT-MR image alignment for both cases.

Influence of family history on risk of second primary cancers and survival in patients with squamous cell skin cancer

BACKGROUND

Patients with squamous cell skin cancer (SCC) have an excellent prognosis but second primary cancers (SPCs) weaken survival prospects. Family history is a known risk factor for cancer but whether it is a risk factor for SPC in patients with SCC is not known.

OBJECTIVES

To quantify the risk of family history on SPCs in patients with SCC and estimate survival probabilities of patients with SPCs depending on family history.

METHODS

With 13 945 histologically verified SCCs, relative risks (RRs) were estimated for family history using a generalized regression model. For survival analysis, hazard ratios (HRs) were assessed using a multivariable Cox proportional-hazards model.

RESULTS

Family history of invasive SCC increased risk of second invasive SCC [RR = 42·92, 95% confidence interval (CI) 33·69-50·32] compared with risk without family history (RR 19·12, 95% CI 17·88-21·08). Family history of any nonskin cancer in invasive SCC increased risk of the same cancers to be diagnosed as SPC (RRF_H = 1·48, 95% CI 1·35-1·61 vs. RR_{no FH} = 1·40, 95% CI 1·32-1·48); significant increases were observed for seven different nonskin cancers. Most results were replicated for in situ SCC. SPC was deleterious for survival irrespective of family history; HR for patients with SPC was 4·28 (95% CI 3·83-4·72) vs. those without SPC (1·04).

CONCLUSIONS

Family history of nonskin cancer was associated with approximately a doubling of risk for SPCs in patients with SCC. SPC increases the death rate in patients with SCC 3-4 times, irrespective of family history. Taking family history into account at SCC diagnosis may help prevention or early detection of SPCs. What's already known about this topic? Second primary cancers (SPCs) are frequently diagnosed in patients with invasive and in situ squamous cell carcinoma (SCC); some epidemiological studies suggest a link to immune dysfunction. Family history of cancer is a risk factor for practically all first primary cancers but whether it also influences risk of SPCs in patients with SCC is not known. The possible influence of family history on survival in patients with SCC remains to be established. Linked Comment: Youlden and Baade. Br J Dermatol 2020; 183:414-415.

Evaluation of an intensity-based algorithm for 2D/3D registration of natural knee videofluoroscopy data

The accurate quantification of in-vivo tibio-femoral kinematics is essential for understanding joint functionality, but determination of the 3D pose of bones from 2D single-plane fluoroscopic images remains challenging. We aimed to evaluate the accuracy, reliability and repeatability of an intensity-based 2D/3D registration algorithm. The accuracy was evaluated using fluoroscopic images of 2 radiopaque bones in 18 different poses, compared against a gold-standard fiducial calibration device. In addition, 3 natural femora and 3 natural tibiae were used to examine registration reliability and repeatability. Both manual fitting and intensity-based registration exhibited a mean absolute error of <1 mm in-plane. Overall, intensity-based registration of the femoral bone model revealed significantly higher translational and rotational errors than manual fitting, while no statistical differences (except for y-axis translation) were found for the tibial bone model. The repeatability of 108 intensity-based registrations showed mean in-plane standard deviations of 0.23-0.56 mm, but out-of-plane position repeatability was lower (mean SD: femur 7.98 mm, tibia 6.96 mm). SDs for rotations averaged 0.77-2.52°. While the algorithm registered some images extremely well, other images clearly required manual intervention. When the algorithm registered the bones repeatably, it was also accurate, suggesting an approach that includes manual intervention could become practical for efficient and accurate registration.

Effect of pelvic tilt and rotation on cup orientation in standing anteroposterior radiographs

BACKGROUND

Individual pelvic tilt and rotation have wide variability that can affect the measurement of cup orientation in anteroposterior (AP) radiographs. The purpose of this study was to analyse the effect of pelvic tilt and rotation on radiographic measurements of cup orientation.

METHODS

A total of 53 patients (63 hips) were included in this study. The patients underwent a computed tomography study with standing AP pelvis radiographs taken both preoperatively and approximately 3 months postoperatively. We used 2-dimensional/3-dimensional matching to measure the pelvic tilt and rotation, and the non-standardised and standardised cup orientation.

RESULTS

There was no difference in the pelvic tilt and rotation between the preoperative and postoperative radiographs. The distribution of the differences between the non-standardised and standardised cup anteversion exhibited a change within 5° in only 34/63 (54%) hips. The pelvic tilt correlated with the difference between the non-standardised and standardised cup anteversion, but the pelvic rotation did not. When all 63 hips were separated into the right and left sides, the pelvic rotation inversely correlated with the pelvic tilt-adjusted difference between the non-standardised and standardised cup anteversion of the right side but directly correlated with that of the left side.

CONCLUSIONS

The current study demonstrated that the measurement of cup anteversion in standing AP radiographs is significantly affected by both the pelvic tilt and pelvic rotation. An improved understanding of the pelvic orientation may eventually allow for desired cup positioning on a patient-specific basis to potentially reduce complications associated with the malposition of the cup.