A fully automatic fiducial detection and correspondence establishing method for online C-arm calibration.

PURPOSE: Online C-arm calibration with a mobile fiducial cage plays an essential role in various image-guided interventions. However, it is challenging to develop a fully automatic approach, which requires not only an accurate detection of fiducial projections but also a robust 2D-3D correspondence establishment. METHODS: We propose a novel approach for online C-arm calibration with a mobile fiducial cage. Specifically, a novel mobile calibration cage embedded with 16 fiducials is designed, where the fiducials are arranged to form 4 line patterns with different cross-ratios. Then, an auto-context-based detection network (ADNet) is proposed to perform an accurate and robust detection of 2D projections of those fiducials in acquired C-arm images. Subsequently, we present a cross-ratio consistency-based 2D-3D correspondence establishing method to automatically match the detected 2D fiducial projections with those 3D fiducials, allowing for an accurate online C-arm calibration. RESULTS: We designed and conducted comprehensive experiments to evaluate the proposed approach. For automatic detection of 2D fiducial projections, the proposed ADNet achieved a mean point-to-point distance of 0.65 ± 1.33 pixels. Additionally, the proposed C-arm calibration approach achieved a mean re-projection error of 1.01 ± 0.63 pixels and a mean point-to-line distance of 0.22 ± 0.12  mm. When the proposed C-arm calibration approach was applied to downstream tasks involving landmark and surface model reconstruction, sub-millimeter accuracy was achieved. CONCLUSION: In summary, we developed a novel approach for online C-arm calibration. Both qualitative and quantitative results of comprehensive experiments demonstrated the accuracy and robustness of the proposed approach. Our approach holds potentials for various image-guided interventions.

Incremental regression of localization context for automatic segmentation of ossified ligamentum flavum from CT data.

PURPOSE: Segmentation of ossified ligamentum flavum (OLF) plays a crucial role in developing computer-assisted, image-guided systems for decompressive thoracic laminectomy. Manual segmentation is time-consuming, tedious, and label-intensive. It also suffers from inter- and intra-observer variability. Automatic segmentation is highly desired. METHODS: A two-stage, localization context-aware framework is developed for automatic segmentation of ossified ligamentum flavum. In the first stage, localization heatmaps of OLFs are obtained via incremental regression. In the second stage, the obtained heatmaps are then treated as the localization context for a segmentation U-Net. Our framework can directly map a whole volumetic data to its volume-wise labels. RESULTS: We designed and conducted comprehensive experiments on datasets of 100 patients to evaluate the performance of the proposed method. Our method achieved an average Dice similarity coefficient of 61.2 ± 7.6%, an average surface distance of 1.1 ± 0.5 mm, and an average positive predictive value of 62.0 ± 12.8%. CONCLUSION: To the best knowledge of the authors, this is the first study aiming for automatic segmentation of ossified ligamentum flavum. Results from the comprehensive experiments demonstrate the superior performance of the proposed method over the state-of-the-art methods.

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.

Glutenin-gliadin ratio changes combining heat-moisture treatment significantly influences the in vitro digestibility of starch in recombinant wheat flours.

This study aimed to investigate the effect of heat-moisture treatment (HMT) on the in vitro digestibility of recombinant wheat flours characterizing by gluten proteins differ in glutenin-gliadin ratio. Compared with the untreated flours in our previous study, HMT improved the digestion resistance of starch in flours with different glutenin-gliadin ratios. For the HMT strong-gluten flour, the proportional increase of glutenin led to an excessively strong and fragile gluten network that unstable under HMT, which weakened the wrapping of gluten network around starch granules and reduced the long- and short-range order of starch, resulting in the conversion of resistant starch (RS) and slowly digestible starch (SDS) to rapidly digestible starch (RDS); however, the quantitative increase of gliadin induced the conversion of SDS to RS due to the enhanced protein-starch interactions as well as the improved long- and short-range order of starch during HMT. For the HMT weak-gluten flour, the changes of glutenin-gliadin ratio aggravated the broken of protein network and starch granules during HMT, thus improving the starch digestibility in varying degrees. In conclusion, the relative crystallinity of starch mainly affected the content of resistant starch, while the content of slowly digestible starch was more influenced by protein-starch interactions.

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 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.

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.

High-Fidelity sEMG Signals Recorded by an on-Skin Electrode Based on AgNWs for Hand Gesture Classification Using Machine Learning.

The human forearm is one of the most densely distributed parts of the human body, with the most irregular spatial distribution of muscles. A number of specific forearm muscles control hand motions. Acquiring high-fidelity sEMG signals from human forearm muscles is vital for human-machine interface (HMI) applications based on gesture recognition. Currently, the most commonly used commercial electrodes for detecting sEMG or other electrophysiological signals have a rigid nature without stretchability and cannot maintain conformal contact with the human skin during deformation, and the adhesive hydrogel used in them to reduce skin-electrode impedance may shrink and cause skin inflammation after long-term use. Therefore, developing elastic electrodes with stretchability and biocompatibility for sEMG signal recording is essential for developing HMI. Here, we fabricated a nanocomposite hybrid on-skin electrode by infiltrating silver nanowires (AgNWs), a one-dimensional (1D) nano metal material with conductivity, into polydimethylsiloxane (PDMS), a silicone elastomer with a similar Young's modulus to that of the human skin. The AgNW on-skin electrode has a thickness of 300 µm and low sheet resistance of 0.481 ± 0.014 Ω/sq and can withstand the mechanical strain of up to 54% and maintain a sheet resistance lower than 1 Ω/sq after 1000 dynamic strain cycles. The AgNW on-skin electrode can record high signal-to-noise ratio (SNR) sEMG signals from forearm muscles and can reflect various force levels of muscles by sEMG signals. Besides, four typical hand gestures were recognized by the multichannel AgNW on-skin electrodes with a recognition accuracy of 92.3% using machine learning method. The AgNW on-skin electrode proposed in this study has great potential and promise in various HMI applications that employ sEMG signals as control signals.

Effects of HMW-GSs at Glu-B1 locus on starch-protein interaction and starch digestibility during thermomechanical processing of wheat dough.

BACKGROUND: The composition of glutenin protein significantly affects protein-starch interactions and starch digestion characteristics in wheat dough matrices. To elucidate the effects of high molecular weight glutenin subunits at the Glu-B1 locus on dough processing quality, the detailed structural changes of protein, starch, and their complexes were compared in Mixolab dough samples of two near isogenic lines 7 + 8 and 7 + 9. RESULTS: The results showed that the degree of protein aggregation increased continuously during dough processing, as did the destruction and rearrangement of the gluten network. Compared to 7 + 8, the stronger and more stable protein network formed in 7 + 9 dough induced intensive interactions between protein and starch, primarily through hydrogen bonds and isomeric glycosidic bonds. In 7 + 9 dough, the more compact and extensive protein-starch network significantly inhibited starch gelatinization during dough pasting, while during the dough cooling stage [from C4 (82.8 °C) to C5 (52.8 °C)], more protein-starch complexes composed of monomeric proteins and short-chain starch were generated, which remarkably inhibited starch retrogradation. All protein-starch interactions in the 7 + 9 dough improved the starch digestion resistance, as reflected by the high content of resistant starch. CONCLUSION: The more extensive and intensive protein-starch interactions in the 7 + 9 dough inhibited the gelatinization and enzymatic hydrolysis of starch, thereby producing more slowly digestible starch and resistant starch. These findings demonstrate the feasibility of optimizing the texture and digestibility of wheat-based food products by regulating the behavior and interactions of proteins and starch during dough processing. © 2022 Society of Chemical Industry.

Pre-baking-steaming of oat induces stronger macromolecular interactions and more resistant starch in oat-buckwheat noodle.

Pre-hydrothermal treatment is widely used to improve the quality of oat or buckwheat noodles. Noodle preparations containing pre-baked-steamed oat and untreated buckwheat (BUN) exhibited the highest sensory score (45.2) among six oat-buckwheat noodle preparations produced with different pre-hydrothermal treatments. Further comparison between the BUN and the noodle prepared with untreated oat and buckwheat showed that, the pre-baking-steaming of oat not only produced a more extended and thermally stable protein network involving rearranged gluten and oat globulins in cooked BUN, but also enhanced the short-range molecular order of gelatinized starch through the formation of nascent double helixes and binary/ternary complexes (starch-lipid, starch-protein and starch-lipid-protein). Overall, these stronger macromolecular interactions in cooked BUN led to an extensive and compact protein-starch network that promoted the formation of more resistant starch (41%). Our findings elucidated the molecular mechanism that underpin the positive effect of oat pretreatment on noodle quality and digestibility.

Effects of Composition and Strength of Wheat Gluten on Starch Structure, Digestion Properties and the Underlying Mechanism.

To understand the effect of gluten on starch digestion characteristics, the structural characteristics of protein, starch, and starch digestion attributes were explored by using flours of four wheat near-isogenic lines. Protein and starch fractions from the four flours were used to form so-called recombinant flours where glutenin and gliadin protein fractions, in different ratios, were combined with starch and heated in a water slurry at 80 °C for 5 min. We found that starch digestibility of the recombinant flours could be reproducibly modified by altering the long- and short-range molecular order of starch through varying the attributes of the gluten protein by virtue of the gluten strength as well as the proportions of glutenin and gliadins. The gluten composition changes of strong-gluten flour did not improve the starch digestion resistibility, however, for the moderate- and weak-gluten flours, the proportional increase of glutenin improved the resistance of starch to digestion through the increased long- and short-range molecular order of starch. The resistance of starch to digestion could also be enhanced with increasing gliadin, and was associated with the modified short-range molecular order of starch. We propose that flour mixtures can be optimized for specified product quality by manipulating the amounts of both gliadin and glutenin.

Dynamic behaviors of protein and starch and interactions associated with glutenin composition in wheat dough matrices during sequential thermo-mechanical treatments.

To clarify the detailed behaviors of protein, starch and interactions during complex dough processing, structural changes in dough protein and starch during continuous Mixolab processing were investigated using wheat near-isogenic lines carrying high-molecular-weight glutenin subunits 1Dx5 + 1Dy10 (5 + 10) or 1Dx2 + 1Dy12 (2 + 12) at the Glu-D1 locus. A more stable gluten network including disulfide bonds and hydrophobic interactions, was formed in the 5 + 10 dough before dough weakening at 53.5 °C, than in the 2 + 12 dough. Thereafter, thermo-mechanical treatment induced the depolymerization of gluten until starch gelatinization peak at 74.6 °C; however, from the peak to trough viscosity at 82.8 °C, additional monomeric proteins were incorporated into the repolymerized proteins characterized by increased disulfide bonds, hydrogen bonds, and ß-sheets. Generally, the protein aggregates of 5 + 10 showed a higher degree of polymerization and better stability than those of 2 + 12 during dough processing, which significantly slowed starch gelatinization and recyclization. Moreover, stronger interactions between monomeric proteins and amylose/short-branch starch via glycosidic and hydrogen bonds were found in 5 + 10 dough during starch pasting and retrogradation. The findings demonstrate the feasibility of optimizing the texture and digestibility of wheat-based food products by regulating the behaviors and interactions of proteins and starch during dough processing.

Macromolecular networks interactions in wheat flour dough matrices during sequential thermal-mechanical treatment.

Differences in Mixolab measurements of dough processing were examined using, as a base, flour from pure breeding, isogenic, wheat lines carrying either the high molecular weight glutenin subunits 5 + 10 or 2 + 12. Before dough pasting, subunits 5 + 10 tend to form a stable gluten network relying mainly on disulfide bonds and hydrogen bonds, but 2 + 12 flour was prone to generating fragile protein aggregates dominated by disulfide bonds and hydrophobicity. During dough pasting, a broader protein network rich in un-extractable polymeric proteins, disulfide bonds and ß-sheets was formed in the dough with subunits 5 + 10, thus resulting in an extensive and compact protein-starch complex which was characterized by high thermal stability and low starch gelatinization, while in the dough of the 2 + 12 line, a porous protein-starch gel with fragmented protein aggregates was controlled by the combination of disulfide bonds, hydrophobicity and hydrogen bonds that facilitated the formation of antiparallel ß-sheets.

Yields and Nutritional of Greenhouse Tomato in Response to Different Soil Aeration Volume at two depths of Subsurface drip irrigation.

This study investigated the effects of 4 aeration levels (varied by injection of air to the soil through subsurface irrigation lines) at two subsurface irrigation line depths (15 and 40 cm) on plant growth, yield and nutritional quality of greenhouse tomato. In all experiments, fruit number, width and length, yield, vitamin C, lycopene and sugar/acid ratio of tomato markedly increased in response to the aeration treatments. Vitamin C, lycopene, and sugar/acid ratio increased by 41%, 2%, and 43%, respectively, in the 1.5 times standard aeration volume compared with the no-aeration treatment. An interaction between aeration level and depth of irrigation line was also observed with yield, fruit number, fruit length, vitamin C and sugar/acid ratio of greenhouse tomato increasing at each aeration level when irrigation lines were placed at 40 cm depth. However, when the irrigation lines were 15 cm deep, the trend of total fruit yields, fruit width, fruit length and sugar/acid ratio first increased and then decreased with increasing aeration level. Total soluble solids and titrable acid decreased with increasing aeration level both at 15 and 40 cm irrigation line placement. When all of the quality factors, yields and economic benefit are considered together, the combination of 40 cm line depth and "standard" aeration level was the optimum combination.

[Effect of biochar addition on soil evaporation.] / 生物炭对土壤水分蒸发的影响.

In order to determine the rational amount of biochar application and its effect on soil hydrological processes in arid area, soil column experiments were conducted in the laboratory using three biochar additions (5%, 10% and 15%) and four different biochar types (d<0.25 mm bamboo charcoal, 0.25 mm

[Pattern recognition of surface electromyography signal based on multi-scale fuzzy entropy].

Action surface electromyography (SEMG) signals can be acquired from human skin surface. Its pattern recognition plays a very important role in practical applications such as human prosthesis and human-computer interface systems. For the purpose of increasing the recognition accuracy, we proposed a new recognition method combining fuzzy entropy (FuzzyEn) with multi-scale analysis. Considering the nonlinear and non-stationary characteristics of the SEMG, a multi-scale fuzzy entropy (MSFuzzyEn) feature was introduced and applied to the pattern recognition of six type action SEMG signals of the forearm. Firstly, multi-scale decomposition was applied to original signal using wavelet decomposition. Then MSFuzzyEn of the decomposed signals were calculated and inputted to support vector machine (SVM) for classification as feature vectors. The mean recognition accuracy reached 97%, which was 3% greater than that when FuzzyEn of original signal is applied to the classification of SEMG signals. The results have proved that the MSFuzzyEn is effective and precise in the classification of action SEMG signals.

[The clinical study of recombinant human thrombopoietin in the treatment of chemotherapy-induced severe thrombocytopenia].

OBJECTIVE: To evaluate the efficacy and safety of recombinant human thrombopoietin (rhTPO) on chemotherapy-induced severe thrombocytopenia. METHODS: In this self-controlled multi-center clinical trial, 81 patients, 23 with solid tumor and 58 with leukemia with complete remission, with the platelet count < or = 20 x 10(9)/L after chemotherapy were given two cycles of the same chemotherapy. The first cycle was non- rhTPO-treated cycle as control, in the second cycle rhTPO of the dosage of 1.0 microg.kg(-1).d(-1) was administered subcutaneously 6-24 hours after the beginning of chemotherapy for at most 14 days. Laboratory tests including complete blood counts, urinalysis, serum chemistry, coagulant test, chest radiography, and electrocardiography were made. Serum samples were screened for anti-rhTPO antibodies. RESULTS: In rhTPO-treated cycle, the platelet count was higher [the mean minimal platelet count was 13 x 10(9)/L, significantly higher than that of the control cycle (12 x 10(9)/L, P = 0.002), the mean maximal platelet count was 186 x 10(9) cells/L, significantly higher than that of the control cycle (122 x 10(9)/L, P < 0.001)]. The duration of thrombocytopenia was shorter in the rhTPO-treated cycle than in the control cycle: days with platelet count <50 x 10(9)/L, days with platelet count recovered > or = 75 x 10(9)/L, and days with platelet count recovered > or = 100 x 10(9)/L were 11 days, 21 days, and 24 days respectively, all significantly shorter than those of the control cycle (13 days, 24 days, and 27 days respectively, P < 0.05, P < 0.001, and P < 0.001). The amount of needed platelet transfusion was 10 U in the rhTPO-treated cycle, both significantly less than those in the control cycle (12 U, P < 0.001). No effects of rhTPO on hemoglobin, white blood cells, hepatic function, kidney function and coagulant function were found. Transient low-titer antibody was developed in one patient. Side effects such as fever, knee arthralgia, dizziness, headache and chill were mild and tolerable. CONCLUSION: Administration of rhTPO after chemotherapy significantly reduces the degree and duration of thrombocytopenia and the need for platelet transfusions.