The Role of Interventional Radiology in the Diagnosis and Treatment of Pulmonary Arteriovenous Malformations.

Pulmonary arteriovenous malformations (PAVMs) are uncommon, predominantly congenital direct fistulous connections between the pulmonary arteries and pulmonary veins, resulting in a right to left shunt. Patients with PAVMs are usually asymptomatic with lesions detected incidentally when radiological imaging is performed for other indications. In this review, we discuss the classification and radiological features of PAVMs as well as their treatment and follow-up options, with a particular focus on percutaneous endovascular techniques and the evolution of the available equipment for treatment.

Detection of Pneumothorax with Deep Learning Models: Learning From Radiologist Labels vs Natural Language Processing Model Generated Labels.

RATIONALE AND OBJECTIVES: To compare the performance of pneumothorax deep learning detection models trained with radiologist versus natural language processing (NLP) labels on the NIH ChestX-ray14 dataset. MATERIALS AND METHODS: The ChestX-ray14 dataset consisted of 112,120 frontal chest radiographs with 5302 positive and 106, 818 negative labels for pneumothorax using NLP (dataset A). All 112,120 radiographs were also inspected by 4 radiologists leaving a visually confirmed set of 5,138 positive and 104,751 negative for pneumothorax (dataset B). Datasets A and B were used independently to train 3 convolutional neural network (CNN) architectures (ResNet-50, DenseNet-121 and EfficientNetB3). All models' area under the receiver operating characteristic curve (AUC) were evaluated with the official NIH test set and an external test set of 525 chest radiographs from our emergency department. RESULTS: There were significantly higher AUCs on the NIH internal test set for CNN models trained with radiologist vs NLP labels across all architectures. AUCs for the NLP/radiologist-label models were 0.838 (95%CI:0.830, 0.846)/0.881 (95%CI:0.873,0.887) for ResNet-50 (p = 0.034), 0.839 (95%CI:0.831,0.847)/0.880 (95%CI:0.873,0.887) for DenseNet-121, and 0.869 (95%CI: 0.863,0.876)/0.943 (95%CI: 0.939,0.946) for EfficientNetB3 (p ≤0.001). Evaluation with the external test set also showed higher AUCs (p <0.001) for the CNN models trained with radiologist versus NLP labels across all architectures. The AUCs for the NLP/radiologist-label models were 0.686 (95%CI:0.632,0.740)/0.806 (95%CI:0.758,0.854) for ResNet-50, 0.736 (95%CI:0.686, 0.787)/0.871 (95%CI:0.830,0.912) for DenseNet-121, and 0.822 (95%CI: 0.775,0.868)/0.915 (95%CI: 0.882,0.948) for EfficientNetB3. CONCLUSION: We demonstrated improved performance and generalizability of pneumothorax detection deep learning models trained with radiologist labels compared to models trained with NLP labels.

Deep Learning Systems for Pneumothorax Detection on Chest Radiographs: A Multicenter External Validation Study.

PURPOSE: To assess the generalizability of a deep learning pneumothorax detection model on datasets from multiple external institutions and examine patient and acquisition factors that might influence performance. MATERIALS AND METHODS: In this retrospective study, a deep learning model was trained for pneumothorax detection by merging two large open-source chest radiograph datasets: ChestX-ray14 and CheXpert. It was then tested on six external datasets from multiple independent institutions (labeled A-F) in a retrospective case-control design (data acquired between 2016 and 2019 from institutions A-E; institution F consisted of data from the MIMIC-CXR dataset). Performance on each dataset was evaluated by using area under the receiver operating characteristic curve (AUC) analysis, sensitivity, specificity, and positive and negative predictive values, with two radiologists in consensus being used as the reference standard. Patient and acquisition factors that influenced performance were analyzed. RESULTS: The AUCs for pneumothorax detection for external institutions A-F were 0.91 (95% CI: 0.88, 0.94), 0.97 (95% CI: 0.94, 0.99), 0.91 (95% CI: 0.85, 0.97), 0.98 (95% CI: 0.96, 1.0), 0.97 (95% CI: 0.95, 0.99), and 0.92 (95% CI: 0.90, 0.95), respectively, compared with the internal test AUC of 0.93 (95% CI: 0.92, 0.93). The model had lower performance for small compared with large pneumothoraces (AUC, 0.88 [95% CI: 0.85, 0.91] vs AUC, 0.96 [95% CI: 0.95, 0.97]; P = .005). Model performance was not different when a chest tube was present or absent on the radiographs (AUC, 0.95 [95% CI: 0.92, 0.97] vs AUC, 0.94 [95% CI: 0.92, 0.05]; P > .99). CONCLUSION: A deep learning model trained with a large volume of data on the task of pneumothorax detection was able to generalize well to multiple external datasets with patient demographics and technical parameters independent of the training data.Keywords: Thorax, Computer Applications-Detection/DiagnosisSee also commentary by Jacobson and Krupinski in this issue.Supplemental material is available for this article.©RSNA, 2021.

Interface Engineering to Eliminate Hysteresis of Carbon-Based Planar Heterojunction Perovskite Solar Cells via CuSCN Incorporation.

A carbon electrode with low cost and high stability exhibited competitiveness for its practical application in organic-inorganic hybrid perovskite solar cells (PSCs). Nonetheless, issues such as poor interface contact with an adjacent perovskite layer and obvious hysteresis phenomenon are bottlenecks that need to be overcome to make carbon-based PSCs (C-PSCs) more attractive in practice. Herein, we report an effective method to enhance the interfacial charge transport of C-PSCs by introducing the CuSCN material into the device. Two types of CuSCN-assisted devices were studied in this work. One was based on the deposition of an ultrathin CuSCN layer between the perovskite absorber layer and the carbon cathode (PSK/CuSCN/C), and the other was by infiltrating CuSCN solution into the carbon film (PSK/C-CuSCN) by taking advantage of the macroporous structure of the carbon. We have found that the CuSCN incorporation by both methods can effectively address the hysteretic feature in planar C-PSCs. The origin for the hysteresis evolution was unraveled by the investigation of the energy alignment and the kinetics of interfacial charge transfer and hole trap-state density. The results have shown that both types of CuSCN-containing devices showed improved interfacial charge carrier extraction, suppressed carrier recombination, reduced trap-state density, and enhanced charge transport, leading to negligible hysteresis. Furthermore, the CuSCN-incorporated C-PSCs demonstrated enhanced device stability. The power conversion efficiency remained 98 and 91% of the initial performance (13.6 and 13.4%) for PSK/CuSCN/C and PSK/C-CuSCN, respectively, after being stored under a high humidity (75-85%) environment for 10 days. The devices also demonstrated extraordinary long-term stability with a negligible performance drop after being stored in air (relative humidity: 33-35%) for 90 days.

Optimization of Mo/Cr bilayer back contacts for thin-film solar cells.

Molybdenum (Mo) is the most commonly used material as back contact in thin-film solar cells. Adhesion of Mo film to soda-lime glass (SLG) substrate is crucial to the performance of solar cells. In this study, an optimized bilayer structure made of a thin layer of Mo on an ultra-thin chromium (Cr) adhesion layer is used as the back contact for a copper zinc tin sulfide (CZTS) thin-film solar cell on a SLG substrate. DC magnetron sputtering is used for deposition of Mo and Cr films. The conductivity of Mo/Cr bilayer films, their microstructure and surface morphology are studied at different deposition powers and working pressures. Good adhesion to the SLG substrate has been achieved by means of an ultra-thin Cr layer under the Mo layer. By optimizing the deposition conditions we achieved low surface roughness, high optical reflectance and low sheet resistivity while we could decrease the back contact thickness to 600 nm. That is two thirds to half of the thickness that is currently being used for bilayer and single layer back contact for thin-film solar cells. We demonstrate the excellent properties of Mo/Cr bilayer as back contact of a CZTS solar cell.

Hindered Formation of Photoinactive δ-FAPbI3 Phase and Hysteresis-Free Mixed-Cation Planar Heterojunction Perovskite Solar Cells with Enhanced Efficiency via Potassium Incorporation.

Organic-inorganic hybrid lead halide perovskite solar cells have demonstrated competitive power conversion efficiency over 22%; nevertheless, critical issues such as unsatisfactory device stability, serious current-voltage hysteresis, and formation of photo nonactive perovskite phases are obstacles for commercialization of this photovoltaics technology. Herein we report a facial yet effective method to hinder formation of photoinactive δ-FAPbI3 and hysteresis behavior in planar heterojunction perovskite solar cells based on K x(MA0.17FA0.83)1- xPbI2.5Br0.5 (0≤ x ≤ 0.1) through incorporation of potassium ions (K+). X-ray diffraction patterns demonstrate formation of photoinactive δ-FAPbI3 was almost completely suppressed after K+ incorporation. Density functional theory calculation shows K+ prefers to enter the interstitial sites of perovskite lattice, leading to chemical environmental change in the crystal structure. Ultrafast transient absorption spectroscopy has revealed that K+ incorporation leads to enhanced carrier lifetime by 50%, which is also confirmed by reduced trap-assisted recombination of the perovskite solar cells containing K+ in photovoltage decay. Ultraviolet photoelectron spectroscopy illustrates that K+ incorporation results in a significant rise of conduction band minimum of the perovskite material by 130 meV, leading to a more favorable energy alignment with electron transporting material. At the optimal content of 3% K+ (molar ratio, relative to the total monovalent cations), nearly hysteresis-free, enhanced power conversion efficiencies from 15.72% to 17.23% were obtained in this solar cell.

Animal models in plastic and reconstructive surgery simulation-a review.

BACKGROUND: The use of live and cadaveric animal models in surgical training is well established as a means of teaching and improving surgical skill in a controlled setting. We aim to review, evaluate, and summarize the models published in the literature that are applicable to Plastic Surgery training. MATERIALS AND METHODS: A PubMed search for keywords relating to animal models in Plastic Surgery and the associated procedures was conducted. Animal models that had cross over between specialties such as microsurgery with Neurosurgery and pinnaplasty with ear, nose, and throat surgery were included as they were deemed to be relevant to our training curriculum. A level of evidence and recommendation assessment was then given to each surgical model. RESULTS: Our review found animal models applicable to plastic surgery training in four major categories namely-microsurgery training, flap raising, facial surgery, and hand surgery. Twenty-four separate articles described various methods of practicing microsurgical techniques on different types of animals. Fourteen different articles each described various methods of conducting flap-based procedures which consisted of either local or perforator flap dissection. Eight articles described different models for practicing hand surgery techniques. Finally, eight articles described animal models that were used for head and neck procedures. CONCLUSIONS: A comprehensive summary of animal models related to plastic surgery training has been compiled. Cadaveric animal models provide a readily available introduction to many procedures and ought to be used instead of live models when feasible.

Precisely Controlled Synthesis of High Quality Kesterite Cu2ZnSnS4 Thin Film via Co-Electrodeposited CuZnSn Alloy Film.

In this work, a facile co-electrodeposition method was used to fabricate CuZnSn alloy films where the content of copper, zinc and tin could be precisely controlled through manipulating the mass transfer process in the electrochemical deposition. By finely tuning the concentration of the cations of Cu2+, Zn2+ and Sn2+ in the electrochemical bath solution, uniform CuZnSn film with desired composition of copper poor and zinc rich was made. Sulphurisation of the CuZnSn alloy film led to the formation of compact and large grains Cu2ZnSnS4 thin film absorber with an optimum composition of Cu/(Zn+Sn) ≈ 0.8, Zn/Sn ≈ 1.2. Both SEM morphology and EDS mapping results confirmed the uniformity of the CuZnSn and Cu2ZnSnS4 films and the homogeneous distribution of Cu, Zn, Sn and S elements in the bulk films. The XRD and Raman measurements indicated that the synthesized Cu2ZnSnS4 film was kesterite phase without impurities detected. Photoelectrochemical tests were carried out to evaluate the CZTS film's photocurrent response under illumination of green light.

One-step synthesis of high quality kesterite Cu2ZnSnS4 nanocrystals - a hydrothermal approach.

The present work demonstrates a systematic approach for the synthesis of pure kesterite-phase Cu2ZnSnS4 (CZTS) nanocrystals with a uniform size distribution by a one-step, thioglycolic acid (TGA)-assisted hydrothermal route. The formation mechanism and the role of TGA in the formation of CZTS compound were thoroughly studied. It has been found that TGA interacted with Cu(2+) to form Cu(+) at the initial reaction stage and controlled the crystal-growth of CZTS nanocrystals during the hydrothermal reaction. The consequence of the reduction of Cu(2+) to Cu(+) led to the formation Cu2- x S nuclei, which acted as the crystal framework for the formation of CZTS compound. CZTS was formed by the diffusion of Zn(2+) and Sn(4+) cations to the lattice of Cu2- x S during the hydrothermal reaction. The as-synthesized CZTS nanocrystals exhibited strong light absorption over the range of wavelength beyond 1000 nm. The band gap of the material was determined to be 1.51 eV, which is optimal for application in photoelectric energy conversion device.

Electrochemically exfoliated graphene for electrode films: effect of graphene flake thickness on the sheet resistance and capacitive properties.

We present an electrochemical exfoliation method to produce controlled thickness graphene flakes by ultrasound assistance. Bilayer graphene flakes are dominant in the final product by using sonication during the electrochemical exfoliation process, while without sonication the product contains a larger percentage of four-layer graphene flakes. Graphene sheets prepared by using the two procedures are processed into films to measure their respective sheet resistance and optical transmittance. Solid-state electrolyte supercapacitors are made using the two types of graphene films. Our study reveals that films with a higher content of multilayer graphene flakes are more conductive, and their resistance is more easily reduced by thermal annealing, making them suitable as transparent conducting films. The film with higher content of bilayer graphene flakes shows instead higher capacitance when used as electrode in a supercapacitor.