[Mitigating metal artifacts from cobalt-chromium alloy crowns in cone-beam CT images through deep learning techniques]

Objective: To develop and evaluate metal artifact removal systems (MARS) based on deep learning to assess their effectiveness in removing artifacts caused by different thicknesses of metals in cone-beam CT (CBCT) images. Methods: A full-mouth standard model (60 mm×75 mm×110 mm) was three-dimensional (3D) printed using photosensitive resin. The model included a removable and replaceable target tooth position where cobalt-chromium alloy crowns with varying thicknesses were inserted to generate matched CBCT images. The artifacts resulting from cobalt-chromium alloys with different thicknesses were evaluated using the structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR). CNN-MARS and U-net-MARS were developed using a convolutional neural network and U-net architecture, respectively. The effectiveness of both MARSs were assessed through visualization and by measuring SSIM and PSNR values. The SSIM and PSNR values were statistically analyzed using one-way analysis of variance (α=0.05). Results: Significant differences were observed in the range of artifacts produced by different thicknesses of cobalt-chromium alloys (all P<0.05), with 1 mm resulting in the least artifacts. The SSIM values for specimens with thicknesses of 1.0, 1.5, and 2.0 mm were 0.916±0.019, 0.873±0.010, and 0.833±0.010, respectively (F=447.89, P<0.001). The corresponding PSNR values were 20.834±1.176, 17.002±0.427, and 14.673±0.429, respectively (F=796.51, P<0.001). After applying CNN-MARS and U-net-MARS to artifact removal, the SSIM and PSNR values significantly increased for images with the same thickness of metal (both P<0.05). When using the CNN-MARS for artifact removal, the SSIM values for 1.0, 1.5 and 2.0 mm were 0.938±0.023, 0.930±0.029, and 0.928±0.020 (F=2.22, P=0.112), while the PSNR values were 30.938±1.495, 30.578±2.154 and 30.553±2.355 (F=0.54, P=0.585). When using the U-net-MARS for artifact removal, the SSIM values for 1.0, 1.5 and 2.0 mm were 0.930±0.024, 0.932±0.017 and 0.930±0.012 (F=0.24, P=0.788), and the PSNR values were 30.291±0.934, 30.351±1.002 and 30.271±1.143 (F=0.07, P=0.929). No significant differences were found in SSIM and PSNR values after artifact removal using CNN-MARS and U-net-MARS for different thicknesses of cobalt-chromium alloys (all P>0.05). Visualization demonstrated a high degree of similarity between the images before and after artifact removal using both MARS. However, CNN-MARS displayed clearer metal edges and preserved more tissue details when compared with U-net-MARS. Conclusions: Both the CNN-MARS and U-net-MARS models developed in this study effectively remove the metal artifacts and enhance the image quality. CNN-MARS exhibited an advantage in restoring tissue structure information around the artifacts compared to U-net-MARS.

[Mitigating metal artifacts in cone-beam CT images through deep learning techniques]

Objective: To develop and evaluate metal artifact removal systems (MARSs) based on deep learning to assess their effectiveness in removing artifacts caused by different thicknesses of metals in cone-beam CT (CBCT) images. Methods: A full-mouth standard model (60 mm×75 mm×110 mm) was three-dimensional (3D) printed using photosensitive resin. The model included a removable and replaceable target tooth position where cobalt-chromium alloy crowns with varying thicknesses were inserted to generate matched CBCT images. The artifacts resulting from cobalt-chromium alloys with different thicknesses were evaluated using the structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR). CNN-MARS and U-net-MARS were developed using a convolutional neural network and U-net architecture, respectively. The effectiveness of both MARSs were assessed through visualization and by measuring SSIM and PSNR values. The SSIM and PSNR values were statistically analyzed using one-way analysis of variance (α=0.05). Results: Significant differences were observed in the range of artifacts produced by different thicknesses of cobalt-chromium alloys (all P<0.05), with 1 mm resulting in the least artifacts. The SSIM values for specimens with thicknesses of 1.0 mm, 1.5 mm, and 2.0 mm were 0.916±0.019, 0.873±0.010, and 0.833±0.010, respectively (F=447.89, P<0.001). The corresponding PSNR values were 20.834±1.176, 17.002±0.427, and 14.673±0.429, respectively (F=796.51, P<0.001). After applying CNN-MARS and U-net-MARS to artifact removal, the SSIM and PSNR values significantly increased for images with the same thickness of metal (both P<0.05). When using the CNN-MARS for artifact removal, the SSIM values for 1.0, 1.5 and 2.0 mm were 0.938±0.023, 0.930±0.029, and 0.928±0.020 (F=2.22, P=0.112), while the PSNR values were 30.938±1.495, 30.578±2.154 and 30.553±2.355 (F=0.54, P=0.585). When using the U-net-MARS for artifact removal, the SSIM values for 1.0, 1.5 and 2.0 mm were 0.930±0.024, 0.932±0.017 and 0.930±0.012 (F=0.24, P=0.788), and the PSNR values were 30.291±0.934, 30.351±1.002 and 30.271±1.143 (F=0.07, P=0.929). No significant differences were found in SSIM and PSNR values after artifact removal using CNN-MARS and U-net-MARS for different thicknesses of cobalt-chromium alloys (all P>0.05). Visualization demonstrated a high degree of similarity between the images before and after artifact removal using both MARSs. However, CNN-MARS displayed clearer metal edges and preserved more tissue details when compared with U-net-MARS. Conclusions: Both the CNN-MARS and U-net-MARS models developed in this study effectively remove the metal artifacts and enhance the image quality. CNN-MARS exhibited an advantage in restoring tissue structure information around the artifacts compared to U-net-MARS.

[Effect of closed extension tube on preventing airway leakage during artificial airway clearance]

Objective: To explore the role of closed extension tube in preventing airway leakage during artificial airway clearance. Methods: The test lung was connected with a ventilator for mechanical ventilation. The pressure parameters were set as 16/5, 20/6, 24/7, 28/8, 32/9 and 36/10 cmH₂O(1 cmH₂O=0.098 kPa), respectively. The circuit was connected with an open extension tube or a closed extension tube. The ventilator was set with different pressure parameters to observe the changes of airway pressure and tidal volume during airway clearance. Results: (1) The pressure parameters were set as 16/5, 20/6, 24/7, 28/8, 32/9 and 36/10 cmH₂O, and the airway pressures (in cmH₂O) of circuit connected with open extension tube were (15.94±0.27)/(4.81±0.04), (20.09±0.23)/(6.05±0.16), (23.89±0.41)/(6.94±0.06), (27.90±0.22)/(7.71±0.18), (31.92±0.13)/(8.74±0.12)and(35.65±0.31)/(9.72±0.07), respectively.Under the same ventilator pressure parameters, the airway pressures (in cmH₂O) of circuit connected with close extension tube were (16.36±0.06)/(4.85±0.04), (20.54±0.26)/(6.44±0.12), (24.36±0.24)/(7.01±0.33), (28.69±0.25)/(8.07±0.08), (32.97±0.33)/(8.93±0.09), (37.34±0.29)/(9.75±0.08), respectively. The airway pressure of circuit connected with open extension tube was lower than that connected with closed extension tube(P<0.05);with the increase of the pressure setting of the ventilator, the difference of the airway pressure between the two extended tubes gradually increased. When the maximum inspiratory pressure of the ventilator was set 36 cmH₂O, the difference reached 1.69 cmH₂O. (2) The airway pressures (in cmH₂O) dropped from (15.94±0.27)/(4.81±0.04), (20.09±0.23)/(6.05±0.16), (23.89±0.41)/(6.94±0.06), (27.90±0.22)/(7.71±0.18), (31.92±0.13)/(8.74±0.12), (35.65±0.31)/(9.72±0.07) to (13.42±0.4)/(3.15±0.14), (16.81±0.6)/(4.30±0.14), (20.22±0.5)/(5.48±0.45), (23.73±1.4)/(6.25±0.22), (24.78±0.7)/(7.13±0.21), (20.83±0.4)/(6.61±0.19)when the suction port of the open extension tube was opened (P<0.05);and the tidal volume (in L) also decreased from 0.328±0.004, 0.580±0.012, 0.621±0.003, 0.626±0.003, 0.615±0.003, 0.603±0.002 to 0.272±0.008, 0.416±0.051, 0.487±0.047, 0.396±0.116, 0.507±0.022, 0.508±0.079, respectively (P<0.05). The decrease of airway pressure and tidal volume gradually increased with the increase of ventilator setting pressure. When the ventilator setting parameter was 36/10 cmH₂O, the decrease of airway inspiratory pressure was (14.82±0.51) cmH₂O and the maximum reduction of tidal volume was (0.164±0.021)L. (3)The airway pressure (in cmH₂O) was increased to(15.70±0.23)/(4.80±0.33), (19.01±0.81)/(5.71±0.34), (22.27±0.62)/(6.85±0.44), (25.35±2.09)/(7.94±0.16), (28.38±0.46)/(8.96±0.23), (33.34±0.71)/(9.71±0.25) when the suction tube was inserted from the suction port of the open extension tube in the open state, and the tidal volume (in L) was increased to 0.340±0.016, 0.563±0.020, 0.571±0.030, 0.556±0.026, 0.514±0.021, 0.512±0.031 as well.The airway pressure and tidal volume of the ventilation circuit were higher than those in the open state, but still lower than those in the closed state. Compared with the closed state of the suction port, the maximum pressure drop and tidal volume decrease were (3.53±0.46) cmH₂O and (0.101±0.011) L, respectively. (4) The pressure of the ventilator was set between 16/5 cmH₂O to 36/10 cmH₂O. The airway pressure (in cmH₂O) was decreased from (16.26±0.04)/(4.85±0.04), (20.74±0.15)/(6.42±0.11), (25.09±0.31)/(7.10±0.13), (29.38±0.24)/(8.17±0.09), (33.80±0.16)/(9.02±0.17), (37.89±0.19)/(9.83±0.07) to(16.36±0.06)/(4.85±0.04), (20.54±0.26)/(6.44±0.12), (24.36±0.24)/(7.01±0.33), (28.69±0.25)/(8.07±0.08), (32.97±0.33)/(8.93±0.09), (37.34±0.29)/(9.75±0.08), respectively during the insertion of the suction tube from the suction port of the closed extension tube, and the tidal volume (in L) was decreased from0.361±0.005, 0.592±0.003, 0.631±0.001, 0.642±0.007, 0.633±0.007, 0.626±0.08 to 0.335±0.005, 0.588±0.008, 0.631±0.002, 0.638±0.004, 0.628±0.004, 0.618±0.005.The maximum pressure change of the ventilation circuit was (0.83±0.27) cm H₂O and the maximum tidal volume change was (0.008±0.006)L. The changes of airway pressure and tidal volume were significantly lower than those of ventilation circuit connected with open extension tube under the same pressure parameters. Conclusion: The connection of closed extension tube in mechanical ventilation circuit can reduce the airway leakage during artificial airway clearance, which is worthy of clinical recommendation.

[Preliminary study on the method of automatically determining facial landmarks based on three-dimensional face template]

Objective: To explore the establishment of an efficient and automatic method to determine anatomical landmarks in three-dimensional (3D) facial data, and to evaluate the effectiveness of this method in determining landmarks. Methods: A total of 30 male patients with tooth defect or dentition defect (with good facial symmetry) who visited the Department of Prosthodontics, Peking University School and Hospital of Stomatology from June to August 2021 were selected, and these participants' age was between 18-45 years. 3D facial data of patients was collected and the size normalization and overlap alignment were performed based on the Procrustes analysis algorithm. A 3D face average model was built in Geomagic Studio 2013 software, and a 3D face template was built through parametric processing. MeshLab 2020 software was used to determine the serial number information of 32 facial anatomical landmarks (10 midline landmarks and 22 bilateral landmarks). Five male patients with no mandibular deviation and 5 with mild mandibular deviation were selected from the Department of Orthodontics or Oral and Maxillofacial Surgery, Peking University School and Hospital of Stomatology from June to August 2021. 3D facial data of patients was collected as test data. Based on the 3D face template and the serial number information of the facial anatomical landmarks, the coordinates of 32 facial anatomical landmarks on the test data were automatically determined with the help of the MeshMonk non-rigid registration algorithm program, as the data for the template method to determine the landmarks. The positions of 32 facial anatomical landmarks on the test data were manually determined by the same attending physician, and the coordinates of the landmarks were recorded as the data for determining landmarks by the expert method. Calculated the distance value of the coordinates of facial anatomical landmarks between the template method and the expert method, as the landmark localization error, and evaluated the effect of the template method in determining the landmarks. Results: For 5 patients with no mandibular deviation, the landmark localization error of all facial anatomical landmarks by template method was (1.65±1.19) mm, the landmark localization error of the midline facial anatomical landmarks was (1.19±0.45) mm, the landmark localization error of bilateral facial anatomical landmarks was (1.85±1.33) mm. For 5 patients with mild mandibular deviation, the landmark localization error of all facial anatomical landmarks by template method was (2.55±2.22) mm, the landmark localization error of the midline facial anatomical landmarks was (1.85±1.13) mm, the landmark localization error of bilateral facial anatomical landmarks was (2.87±2.45) mm. Conclusions: The automatic determination method of facial anatomical landmarks proposed in this study has certain feasibility, and the determination effect of midline facial anatomical landmarks is better than that of bilateral facial anatomical landmarks. The effect of determining facial anatomical landmarks in patients without mandibular deviation is better than that in patients with mild mandibular deviation.

[Long-term outcome of patients with hypertrophic obstructive cardiomyopathy post percutaneous transluminal septal ablation]

Objective: To evaluate the long-term outcome of patients with hypertrophic obstructive cardiomyopathy(HOCM) after percutaneous transluminal septal ablation(PTSMA). Methods: HOCM patients who underwent PTSMA and surgical myectomy at the Chest Hospital of Shanghai Jiao Tong University from April 2001 to February 2019 were included in this retrospective analysis. Patients were divided into PTSMA group and surgical myectomy group. In addition, patients undergoing PTSMA were further divided into HOCM-PTSMA non-survivor group and HOCM-PTSMA survivor group. The general clinical information, procedural/surgical information and complications during hospitalization were compared between groups. Multivariate Cox regression model was used to analyze the independent risk factors for all-cause death in HOCM patients after PTSMA. Results: A total of 104 patients with HOCM who underwent PTSMA were enrolled. Mean age of the patients was (54±15) years old, including 41 females (38.7%). The follow-up time was 37.5(14.3, 76.8) months. At the last follow-up, 12 patients died (HOCM-PTSMA non-survivor group) and 92 were alive(HOCM-PTSMA survivor group). The proportion of patients with NYHA function class Ⅲ/Ⅳ was higher(P=0.036), and the posterior wall of the left ventricle was thicker(P=0.006) in the HOCM-PTSMA non-survivor group than in the HOCM-PTSMA survivor group. The immediate success rate of PTSMA in this cohort was 66%(70/104). The amount of absolute alcohol during the operation in the HOCM-PTSMA non-survivor group was (2.9±0.8) ml, which tended to be higher as compared to that in the HOCM-PTSMA survivor group((2.4±1.0)ml, P=0.056). Kaplan-Meier survival curve analysis showed that patients with HOCM who underwent PTSMA had an all-cause mortality-free survival rate of 90.1%, 78.3%, and 56.9% at 5, 10 and 15 years, and a HOCM-free survival rate of 91.3%, 79.4% and 57.7% at 5, 10 and 15 years, respectively. Multivariate Cox regression analysis showed that age≥ 65 years was an independent risk factor for all-cause death after PTSMA in patients with HOCM (HR=2.697, 95%CI 1.292-18.977, P=0.020). There were 32 patients in the surgical myectomy group. The proportion of patients with NYHA function class Ⅲ/Ⅳ was higher than that in the PTSMA group(P<0.001), while age, gender, and major comorbidities(atrial fibrillation, coronary heart disease, hypertension, and diabetes) as well as the left atrium dimension were all similar between the two groups(all P>0.05). Patients in the surgical myectomy group were followed up for 38.0(17.6, 64.2)months, and no deaths occurred during the follow-up period. Kaplan-Meier survival curve analysis showed that there were no statistically significant differences in all-cause-free and HOCM-free survival rates between patients in PTSMA group and surgical myectomy group(P=0.089 and 0.110, respectively). Conclusion: PTSMA is safe and effective for the treatment of patients with HOCM, and the long-term survival rate of patients after PTSMA is similar as patients undergoing classical surgical myectomy surgery.

[Risk factors associated with prolonged postoperative length of stay of patients with gastric cancer]

Objective: To investigate the risk factors of prolonged postoperative length of stay (LOS) in patients with gastric cancer. Methods: A retrospective study was performed on 2033 patients who underwent curative surgery in Department of Pancreatic-gastric, Cancer Hospital, Chinese Academy of Medical Sciences. Eightieth percentile of postoperative hospital stay for all patients was taken as the cutoff, the patients were divided into the normal group (1 532 patients) and the extended group (501 patients). Logistic regression model was used to determine the risk factors related to prolonged postoperative LOS in gastric cancer. Results: A total of 417 cases with postoperative complications were recruited in extended group. The five highest complications accounting for prolonged LOS were: abdominal infection in 69 cases (13.77%), pleural effusion in 60 cases (11.98%), anastomotic leakage in 43 cases (8.58%), poor wound healing in 34 cases (6.79%), and bleeding in 25 cases (4.99%). Logistic regression analysis showed that age (OR=1.348, P=0.013), surgical procedure (OR=2.712, P<0.001), extent of resection (OR=2.035, P<0.001), degree of incision healing(OR=4.867, P<0.001), and perioperative blood transfusion (OR=1.711, P<0.001) were independent risk factors for prolonged postoperative LOS. Conclusions: The most common complication associated with prolonged postoperative LOS for gastric cancer patients is abdominal infection. Age, surgical procedure, extent of resection, degree of incision healing, and perioperative blood transfusion are the independent risk factors for prolonged postoperative LOS.

Leptocarpin Suppresses Proliferation, Migration, and Invasion of Human Osteosarcoma by Targeting Type-1 Insulin-Like Growth Factor Receptor (IGF-1R)

BACKGROUND Leptocarpin (LTC) has drawn much attention for suppressing tumor growth or reducing inflammation. However, the effect of LTC on osteosarcoma has rarely been reported. Our object was to determine whether LTC suppresses MG63 cell proliferation, migration, and invasion, and whether type-1 insulin-like growth factor receptor (IGF-1R) is one of the targets in LTC suppressing osteosarcoma. MATERIAL AND METHODS Cytotoxicity of LTC was performed by use of a cell-counting kit-8 (CCK-8). RNA interference (RNAi) or pEABE-bleo IGF-1R plasmid were used for silencing or overexpressing IGF-1R, Western blot (WB) analysis was used for IGF-1R expression, CCK-8 for proliferation, and transwell assay for migration and invasion. RESULTS LTC (23.533 μM) treatment for 48 h was taken as the 50% inhibiting concentration (IC50), which significantly (P<0.05) suppressed MG63 cells proliferation, migration, and invasion. LTC (IC50) obviously inhibited IGF-1R expression in MG63 cells, with similar effect to small interfering RNA (siRNA), while pEABE-bleo IGF-1R transfection overexpressed IGF-1R. siRNA silencing IGF-1R suppressed MG63 cells proliferation, migration, and invasion, while pEABE-bleo IGF-1R transfection was significantly (P<0.05) promoted. With or without siRNA or pEABE-bleo IGF-1R transfection, LTC (IC50) suppressed MG63 cells proliferation, migration, and invasion. The effect of LTC (IC50) combined with siRNA on suppressing MG63 cells proliferation, migration, and invasion was more obvious, while the effect of LTC (IC50) combined with pEABE-bleo IGF-1R transfection was less significant (P<0.05). CONCLUSIONS LTC suppressed osteosarcoma proliferation, migration, and invasion by inhibiting IGF-1R expression. IGF-1R is one of the targets in LTC suppressing osteosarcoma.

Isopsoralen-mediated suppression of bone marrow adiposity and attenuation of the adipogenic commitment of bone marrow-derived mesenchymal stem cells

Osteoporosis (OP) increases the risk of bone fractures and other complications, and is thus a major clinical problem. In this study, we examined the effect of isopsoralen on the differentiation of bone-derived marrow mesenchymal stem cells (BMSCs) into osteoblasts and adipocytes, as well as bone formation under osteoporotic conditions. Primary femoral BMSCs isolated from C57BL/6 mice were used to evaluate the isopsoralen-mediated regulation of the expression of alkaline phosphatase (ALP), osteocalcin (OCN) and runt-related transcription factor 2 (RUNX2) during osteogenesis 2 weeks. We also examined the expression of peroxisome proliferator-activated receptor γ (PPARγ) and CCAAT/enhancer binding protein β (C/EBPβ) under adipogenic conditions for 1 and 2 weeks. In addition, ovariectomized (OVX) mice were used to examine the effects of isopsoralen on bone formation for 2 months. Finally, mammalian target of rapamycin complex 1 (mTORC1) signaling was examined under osteogenic and adipogenic conditions. We found that following treatment with isopsoralen, the expression levels of ALP, OCN and RUNX2 were upregulated, whereas those of PPARγ and C/EBPβ were downregulated. mTORC1 signaling was also inhibited in vitro and in vivo. In the OVX mice that were intragastrically administered isopsoralen, bone parameters (trabecular thickness, bone volume/total volume and trabecular number) in the distal femoral metaphysis were significantly increased and the adipocyte number was decreased. On the whole, our findings demonstrate that isopsoralen promoted BMSC differentiation into osteoblasts and suppressed differentiation into adipocytes.

RGD-conjugated iron oxide magnetic nanoparticles for magnetic resonance imaging contrast enhancement and hyperthermia

The purpose of this study was to develop a specific targeting magnetic nanoparticle probe for magnetic resonance imaging and therapy in the form of local hyperthermia. Carboxymethyl dextran-coated ultrasmall superparamagnetic iron oxide nanoparticles with carboxyl groups were coupled to cyclic arginine-glycine-aspartic peptides for integrin α(v)β₃ targeting. The particle size, magnetic properties, heating effect, and stability of the arginine-glycine-aspartic-ultrasmall superparamagnetic iron oxide were measured. The arginine-glycine-aspartic-ultrasmall superparamagnetic iron oxide demonstrates excellent stability and fast magneto-temperature response. Magnetic resonance imaging signal intensity of Bcap37 cells incubated with arginine-glycine-aspartic-ultrasmall superparamagnetic iron oxide was significantly decreased compared with that incubated with plain ultrasmall superparamagnetic iron oxide. The preferential uptake of arginine-glycine-aspartic-ultrasmall superparamagnetic iron oxide by target cells was further confirmed by Prussian blue staining and confocal laser scanning microscopy.