Manual Adjustment Research Articles

Bayesian tensor network structure search and its application to tensor completion

Tensor network (TN) has demonstrated remarkable efficacy in the compact representation of high-order data. In contrast to the TN methods with pre-determined structures, the recently introduced tensor network structure search (TNSS) methods automatically learn a compact TN structure from the data, gaining increasing attention. Nonetheless, TNSS requires time-consuming manual adjustments of the penalty parameters that control the model complexity to achieve better performance, especially in the presence of missing or noisy data. To provide an effective solution to this problem, in this paper, we propose a parameters tuning-free TNSS algorithm based on Bayesian modeling, aiming at conducting TNSS in a fully data-driven manner. Specifically, the uncertainty in the data corruption is well-incorporated in the prior setting of the probabilistic model. For TN structure determination, we reframe it as a rank learning problem of the fully-connected tensor network (FCTN), integrating the generalized inverse Gaussian (GIG) distribution for low-rank promotion. To eliminate the need for hyperparameter tuning, we adopt a fully Bayesian approach and propose an efficient Markov chain Monte Carlo (MCMC) algorithm for posterior distribution sampling. Compared with the previous TNSS method, experiment results demonstrate the proposed algorithm can effectively and efficiently find the latent TN structures of the data under various missing and noise conditions and achieves the best recovery results. Furthermore, our method exhibits superior performance in tensor completion with real-world data compared to other state-of-the-art tensor-decomposition-based completion methods.

A novel decision support system based on computational intelligence and machine learning: Towards zero-defect manufacturing in injection molding

Real-time monitoring solutions have gained popularity across industries due to the advent of Industry 4.0, AI, and big data enhancing the efficiency of industrial production and equipment decisions. Machine learning models that possess computing intelligence and interpretability provide superior predictive capabilities compared to manual adjustments, resulting in cost savings and manufacturing high-quality products. This study proposes a zero-defect manufacturing decision support system based on computational intelligence feature selection combined with interpretable machine learning. The decision support system integrates Particle Swarm Optimization (PSO) and the C4.5 decision tree method, abbreviated as PSO+C4.5, to enable the continuous monitoring of the injection molding process in real-time, considering production parameter information and collected data quality, guiding the decision-making process for implementing zero-defect manufacturing (ZDM). In contrast to existing research, our innovative methodology relies on computational intelligence techniques for extracting features and employs interpretable machine learning prediction models. In terms of quality prediction, our empirical findings show that the suggested method accomplishes the optimal balance between interpretability and predictive performance (Accuracy: 0.9889, Sensitivity: 0.9869, and Specificity: 0.9935). These characteristics can directly support maintenance personnel and operators in optimizing the processing quality process.

Global optimization for the design of shell and tube horizontal thermosyphon reboiler

Shell and tube horizontal thermosyphon reboilers (STHTs) find wide applications in the process industry particularly due to their inherent advantages in handling large duties, high viscosity fluids and low fouled susceptibility. However, current design commercial packages suffer from obvious limitations including extensive user manipulations, trial-and-error necessity, increased human factor influence. To overcome these limitations, we develop in this work a rigorous MINLP model for STHT sizing, by achieving automated selection of detailed exchanger geometries with optimal overall heat transfer coefficient. The corrected minimum cross-flow area in shell-side is incorporated to accommodate full cross-flow patterns in tube bundle, thus handling popular service of X shell without baffle construction. A dependable convective boiling heat transfer model is proposed to obtain the shell-side heat transfer coefficient, validated with Aspen EDR. The critical variable, heat flux, which affects nucleate boiling heat transfer performance, is optimized through a simultaneous equation-solving process using a global solver, reducing excessive unit overdesign. Pressure balance between STHT and distillation column is fully considered by a proposed optimization model incorporating pressure drop and pipe sizing. By employing global optimization techniques and automating the design process, this innovative approach minimizes manual adjustments and provides more accurate and cost-effective designs.

The multimodality cell segmentation challenge: toward universal solutions.

Cell segmentation is a critical step for quantitative single-cell analysis in microscopy images. Existing cell segmentation methods are often tailored to specific modalities or require manual interventions to specify hyper-parameters in different experimental settings. Here, we present a multimodality cell segmentation benchmark, comprising more than 1,500 labeled images derived from more than 50 diverse biological experiments. The top participants developed a Transformer-based deep-learning algorithm that not only exceeds existing methods but can also be applied to diverse microscopy images across imaging platforms and tissue types without manual parameter adjustments. This benchmark and the improved algorithm offer promising avenues for more accurate and versatile cell analysis in microscopy imaging.

AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models

Large Vision-Language Models (LVLMs) such as MiniGPT-4 and LLaVA have demonstrated the capability of understanding images and achieved remarkable performance in various visual tasks. Despite their strong abilities in recognizing common objects due to extensive training datasets, they lack specific domain knowledge and have a weaker understanding of localized details within objects, which hinders their effectiveness in the Industrial Anomaly Detection (IAD) task. On the other hand, most existing IAD methods only provide anomaly scores and necessitate the manual setting of thresholds to distinguish between normal and abnormal samples, which restricts their practical implementation. In this paper, we explore the utilization of LVLM to address the IAD problem and propose AnomalyGPT, a novel IAD approach based on LVLM. We generate training data by simulating anomalous images and producing corresponding textual descriptions for each image. We also employ an image decoder to provide fine-grained semantic and design a prompt learner to fine-tune the LVLM using prompt embeddings. Our AnomalyGPT eliminates the need for manual threshold adjustments, thus directly assesses the presence and locations of anomalies. Additionally, AnomalyGPT supports multi-turn dialogues and exhibits impressive few-shot in-context learning capabilities. With only one normal shot, AnomalyGPT achieves the state-of-the-art performance with an accuracy of 86.1%, an image-level AUC of 94.1%, and a pixel-level AUC of 95.3% on the MVTec-AD dataset.

HybridOctree_Hex: Hybrid octree-based adaptive all-hexahedral mesh generation with Jacobian control

We present a new software package, “HybridOctree_Hex,” for adaptive all-hexahedral mesh generation based on hybrid octree and quality improvement with Jacobian control. The proposed HybridOctree_Hex begins by detecting curvatures and narrow regions of the input boundary to identify key surface features and initialize an octree structure. Subsequently, a strongly balanced octree is constructed using the balancing and pairing rules. Inspired by our earlier preliminary hybrid octree-based work, templates are designed to guarantee an all-hexahedral dual mesh generation directly from the strongly balanced octree. With these pre-defined templates, the sophisticated hybrid octree construction step is skipped to achieve an efficient implementation. After that, elements outside and around the boundary are removed to create a core mesh. The boundary points of the core mesh are connected to their corresponding closest points on the surface to fill the buffer zone and build the final mesh. Coupled with smart Laplacian smoothing, HybridOctree_Hex takes advantage of a delicate optimization-based quality improvement method considering geometric fitting, Jacobian and scaled Jacobian, to achieve a minimum scaled Jacobian that is higher than 0.5. We empirically verify the robustness and efficiency of our method by running the HybridOctree_Hex software on dozens of complex 3D models without any manual intervention or parameter adjustment. We provide the HybridOctree_Hex source code, along with comprehensive results encompassing the input and output files and statistical data in the following repository: https://github.com/CMU-CBML/HybridOctree_Hex.

Feasibility of stereotactic optical navigation for needle positioning in percutaneous nephrolithotomy

BackgroundThis study assessed the feasibility of acquiring single-attempt access to the pelvicalyceal system during percutaneous nephrolithotomy (PCNL) using stereotactic optical navigation combined with cone-beam CT (CBCT) imaging.MethodsPatients with a PCNL indication were prospectively included in this IRB approved study. After sterile preparation, fiducial markers were attached to patients’ skin. An initial intraprocedural CBCT scan was acquired, on which the urologist planned the needle trajectory using the navigation software. After verifying that no critical structures were crossed, the needle guide was aligned with the plan. A needle was manually inserted through the needle guide to the indicated depth and a second CBCT scan was performed for needle position confirmation. Both, scanning and needle insertion, were performed under apnea. The study evaluated technical success, accuracy, procedure time, complication rate, and radiation dose.ResultsBetween June 2022 and April 2023, seven patients were included. In all patients, the navigation system allowed safe puncture. However, the technical success rate was only 29%. In 42% of the cases, pelvicalyceal access was achieved by a small manual adjustment. In the remaining 29%, the needle was retracted and positioned per clinical standard. The average deviation between the needle and target was 5.9 ± 2.3 mm. The average total procedure time was 211 ± 44 min. The average radiation exposure was 6.4 mSv, with CBCT scanning contributing to 82% of this exposure.ConclusionsThe optical navigation system facilitated safe needle insertion but did not consistently ensure accurate one-attempt needle positioning for PCNL. Real-time visualization and trajectory correction may improve the technical success rate.

Achieving Accurate In-Text Citations with EndNote in Academic Writing

EndNote, alongside other citation management software programs like RefWorks, Mendeley, and Zotero, functions as a crucial instrument for archiving, structuring, accessing, and referencing sources in scholarly composition. This study delves into my personal experience of refining EndNote’s default in-text citations across various reference types, for achieving accurate in-text citations adhering to the American Psychological Association (APA) style. To do so, I compiled my personal, albeit disjointed, experiences with refining the default in-text citation format sourced from the EndNote library. The findings are presented and discussed in a thematic manner, focusing on the in-text citation types and how they can be effectively managed within Microsoft (MS) Word documents. The findings ultimately affirm that EndNote’s default in-text citations can be tailored to accurately cite a broader spectrum of citation types. This implies that judicious use of EndNote facilitates precise and systematic citations, with minimal manual adjustments. This consequently streamlines and accelerates the entirety of the course of academic writing.

Clinical acceptance and dosimetric impact of automatically delineated elective target and organs at risk for head and neck MR-Linac patients.

MR-Linac allows for daily online treatment adaptation to the observed geometry of tumor targets and organs at risk (OARs). Manual delineation for head and neck cancer (HNC) patients takes 45-75 minutes, making it unsuitable for online adaptive radiotherapy. This study aims to clinically and dosimetrically validate an in-house developed algorithm which automatically delineates the elective target volume and OARs for HNC patients in under a minute. Auto-contours were generated by an in-house model with 2D U-Net architecture trained and tested on 52 MRI scans via leave-one-out cross-validation. A randomized selection of 684 automated and manual contours (split half-and-half) was presented to an oncologist to perform a blind test and determine the clinical acceptability. The dosimetric impact was investigated for 13 patients evaluating the differences in dosage for all structures. Automated contours were generated in 8 seconds per MRI scan. The blind test concluded that 114 (33%) of auto-contours required adjustments with 85 only minor and 15 (4.4%) of manual contours required adjustments with 12 only minor. Dosimetric analysis showed negligible dosimetric differences between clinically acceptable structures and structures requiring minor changes. The Dice Similarity coefficients for the auto-contours ranged from 0.66 ± 0.11 to 0.88 ± 0.06 across all structures. Majority of auto-contours were clinically acceptable and could be used without any adjustments. Majority of structures requiring minor adjustments did not lead to significant dosimetric differences, hence manual adjustments were needed only for structures requiring major changes, which takes no longer than 10 minutes per patient.

Development of an MR-Guided Focused Ultrasound (MRgFUS) Lesioning Approach for the Fornix in the Rat Brain

Objective: High-intensity magnetic resonance-guided focused ultrasound (MRgFUS) is a noninvasive therapy to lesion brain tissue, used clinically in patients and preclinically in several animal models. Challenges with focused ablation in rodent brains can include skull and near-field heating and accurately targeting small and deep brain structures. We overcame these challenges by creating a novel method consisting of a craniectomy skull preparation, a high-frequency transducer (3 MHz) with a small ultrasound focal spot, a transducer positioning system with an added manual adjustment of ∼0.1 mm targeting accuracy, and MR acoustic radiation force imaging for confirmation of focal spot placement.Methods: The study consisted of two main parts. First, two skull preparation approaches were compared. A skull thinning approach (n=7 lesions) was compared to a craniectomy approach (n=22 lesions), which confirmed a craniectomy was necessary to decrease skull and near-field heating. Second, the two transducer positioning systems were compared with the fornix chosen as a subcortical ablation target. We evaluated the accuracy of targeting using histological methods from a high-frequency transducer with a small ultrasound focal spot and MR acoustic radiation force imaging.Results: Comparing a motorized adjustment system (∼1 mm precision, n=17 lesions) to the motorized system with an added micromanipulator (∼0.1 mm precision, n=14 lesions), we saw an increase in the accuracy of targeting the fornix by 133%. The described work allows for repeatable and accurate targeting of small and deep structures in the rodent brain, such as the fornix, enabling the investigation of neurological disorders in chronic disease models.

Grid-less wideband direction of arrival estimation based on variational Bayesian inference.

Many recent works have addressed the problem of wideband direction of arrival (DOA) estimation using grid-less sparse techniques, and these methods have been shown to outperform the traditional wideband DOA estimation methods. However, these methods often suffer from the problem of requiring manual parameter tuning or high computational complexity, which reduces their practicality. To alleviate this problem, a grid-less wideband DOA estimation method based on variational Bayesian inference is proposed in this paper. The method approximates the posterior probability density function of DOA with the help of variational Bayesian inference, which does not require manual adjustment of parameters and can obtain accurate DOA estimation results with low computational complexity. Numerical simulations and real measurement data processing show that the proposed method has a higher DOA estimation accuracy than other grid-less wideband methods while providing higher computational speed.

Craniofacial phenotyping with fetal MRI: a feasibility study of 3D visualisation, segmentation, surface-rendered and physical models

This study explores the potential of 3D Slice-to-Volume Registration (SVR) motion-corrected fetal MRI for craniofacial assessment, traditionally used only for fetal brain analysis. In addition, we present the first description of an automated pipeline based on 3D Attention UNet trained for 3D fetal MRI craniofacial segmentation, followed by surface refinement. Results of 3D printing of selected models are also presented.Qualitative analysis of multiplanar volumes, based on the SVR output and surface segmentations outputs, were assessed with computer and printed models, using standardised protocols that we developed for evaluating image quality and visibility of diagnostic craniofacial features. A test set of 25, postnatally confirmed, Trisomy 21 fetal cases (24–36 weeks gestational age), revealed that 3D reconstructed T2 SVR images provided 66–100% visibility of relevant craniofacial and head structures in the SVR output, and 20–100% and 60–90% anatomical visibility was seen for the baseline and refined 3D computer surface model outputs respectively. Furthermore, 12 of 25 cases, 48%, of refined surface models demonstrated good or excellent overall quality with a further 9 cases, 36%, demonstrating moderate quality to include facial, scalp and external ears. Additional 3D printing of 12 physical real-size models (20–36 weeks gestational age) revealed good/excellent overall quality in all cases and distinguishable features between healthy control cases and cases with confirmed anomalies, with only minor manual adjustments required before 3D printing.Despite varying image quality and data heterogeneity, 3D T2w SVR reconstructions and models provided sufficient resolution for the subjective characterisation of subtle craniofacial features. We also contributed a publicly accessible online 3D T2w MRI atlas of the fetal head, validated for accurate representation of normal fetal anatomy.Future research will focus on quantitative analysis, optimizing the pipeline, and exploring diagnostic, counselling, and educational applications in fetal craniofacial assessment.

Experience of Implementing Deep Learning-Based Automatic Contouring in Breast Radiation Therapy Planning: Insights From Over 2000 Cases

Background and purposeThis study evaluates the impact and clinical utility of an auto-contouring system for radiation therapy treatments. Materials and methodsThe auto-contouring system was implemented in 2019. We evaluated data from 2,428 patients who underwent adjuvant breast radiation therapy before and after the system's introduction. We collected the treatment's finalized contours, reviewed and revised by multidisciplinary team. After implementation, the treatment contours underwent a finalization process which involved manual review and adjustment of the initial auto-contours. For the pre-implementation group (n = 369), auto-contours were generated retrospectively. We compared the auto-contours and final contours using the Dice similarity coefficient (DSC) and the 95% Hausdorff distance (HD95). ResultsWe analyzed 22,215 structures from final and corresponding auto-contours. The final contours were generally larger, encompassing more slices in the superior or inferior directions. For organs-at-risk (OAR), the heart, esophagus, spinal cord, and contralateral breast demonstrated significantly increased DSC and reduced HD95 post-implementation (all p < 0.05), except for the lungs, which presented inaccurate segmentation. Among target volumes, CTVn_L2, L3, L4, and IMN showed increased DSC and reduced HD95 in the post-implementation group (all p < 0.05), although the increase was less pronounced than OAR outcomes. The analysis also covered factors contributing to significant differences, pattern identification, and outlier detection. ConclusionsIn our study, the adoption of the auto-contouring system was associated with an increased reliance on automated settings, underscoring its utility and the potential risk of automation bias. Given these findings, we believe it may be important to consider the integration of stringent risk assessments and quality management strategies as a precautionary measure for the optimal use of such systems.

Low-Voltage Distribution Network Loss-Reduction Method Based on Load-Timing Characteristics and Adjustment Capabilities

The primary contributors to elevated line losses in low-voltage distribution networks are three-phase load imbalances and variations in load peak–valley differentials. The conventional manual phase sequence adjustment fails to capitalize on the temporal characteristics of the load, and the proliferation of smart homes has opened up new scheduling possibilities for managing the load. Consequently, this paper introduces a loss-reduction method for low-voltage distribution networks that leverages load-timing characteristics and adjustment capabilities. This method combines dynamic and static methods to reduce energy consumption from different time scales. To commence, this paper introduced a hierarchical fuzzy C-means algorithm (H-FCM), taking into account the distance and similarity of load curves. Subsequently, a phase sequence adjustment method, grounded in load-timing characteristics, was developed. The typical user load curve, derived from the classification of user loads, serves as the foundation for constructing a long-term commutation model, therefore mitigating the impact of load fluctuations on artificial commutation. Following this, this paper addressed the interruptible and transferable characteristics of various smart homes. This paper proposed a multi-objective transferable load (TL) optimal timing task adjustment model and a peak-shaving control strategy specifically designed for maximum sustainable power reduction of temperature-controlled loads (TCL). These strategies aim to achieve real-time load adjustment, correct static commutation errors, and reduce peak-to-valley differences. Finally, a simulation verification model was established in MATLAB (R2022a). The results show that the proposed method mainly solves the problems of three-phase imbalance and large load peak–valley difference in low-voltage distribution networks and reduces the line loss of low-voltage distribution networks through manual commutation and load adjustment.

Using automatic model calibration for 3D morphological simulations: a case study of the Bodendorf reservoir flushing

AbstractReservoir sedimentation poses a significant challenge to water resource management. Improving the lifespan and productivity of reservoirs requires appropriate sediment management strategies, among which flushing operations have become more prevalent in practice. Numerical modeling offers a cost-effective approach to assessing the performance of different flushing operations. However, calibrating highly parametrized morphological models remains a complex task due to inherent uncertainties associated with sediment transport processes and model parameters. Traditional calibration methods require laborious manual adjustments and expert knowledge, hindering calibration accuracy and efficiency and becoming impractical when dealing with several uncertain parameters. A solution is to use optimization techniques that enable an objective evaluation of the model behavior by expediting the calibration procedure and reducing the issue of subjectivity. In this paper, we investigate bed level changes as a result of a flushing event in the Bodendorf reservoir in Austria by using a three-dimensional numerical model coupled with an optimization algorithm for automatic calibration. Three different sediment transport formulae (Meyer-Peter and Müller, van Rijn, and Wu) are employed and modified during the calibration, along with the roughness parameter, active layer thickness, volume fraction of sediments in bed, and the hiding-exposure parameter. The simulated bed levels compared to the measurements are assessed by several statistical metrics in different cross-sections. According to the goodness-of-fit indicators, the models using the formulae of van Rijn and Wu outperform the model calculated by the Meyer-Peter and Müller formula regarding bed patterns and the volume of flushed sediments.

Robotic-assisted CT-guided percutaneous thermal ablation of abdominal tumors: An analysis of 41 patients

PurposeRobotic assistance is rapidly evolving and may help physicians optimize needle guidance during percutaneous interventions. The purpose of the study was to report feasibility, safety, accuracy, immediate clinical success and short-term local tumor control after robotic-assisted computed tomography (CT)-guided thermal ablation of abdominal tumors. Materials and methodsForty-one patients who underwent percutaneous thermal ablation of abdominal tumors using robotic-assisted CT-guided were included. All ablations were performed with robotic assistance, using an optically-monitored robotic system with a needle guide sent to preplanned trajectories defined on three-dimensional-volumetric CT acquisitions with respiration monitoring. Endpoints were technical success, safety, distance from needle tip to planned trajectory and number of needle adjustments, and complete ablation rate. ResultsForty-one patients (31 men; mean age, 66.7 ± 9.9 [standard deviation (SD)] years [age range: 41–84 years]) were treated for 48 abdominal tumors, with 79 planned needles. Lesions treated were located in the liver (23/41; 56%), kidney (14/41;34%), adrenal gland (3/41; 7%) or retroperitoneum (1/41; 2%). Technical success was achieved in 39/41 (95%) patients, and 76/79 (96%) needle insertions. The mean lateral distance between the needle tip and planned trajectory was 3.2 ± 4.5 (SD) mm (range: 0–20 mm) before adjustments, and the mean three-dimensional distance was 1.6 ± 2.6 (SD) mm (range: 0–13 mm) after 29 manual depth adjustments (29/78; 37%) and 33 lateral adjustments (33/78; 42%). Two (2/79; 3%) needles required complete manual reinsertion. One grade 3 complication was reported in one patient (1/41; 2%). The overall clinical success rate was 100%. The 3-month local tumor control rate (progression free survival) was 95% (38/41). ConclusionThese results provide further evidence on the use of robotic-assisted needle insertion regarding feasibility, safety, and accuracy, resulting in effective percutaneous thermal ablation of abdominal tumors.

Real-Time Adjustment Method for Metro Systems with Train Delays Based on Improved Q-Learning

This paper presents a solution to address the challenges of unexpected events in the operation of metro trains, which can lead to increased delays and safety risks. An improved Q-learning algorithm is proposed to reschedule train timetables via incorporating train detention and different section running times as actions. To enhance computational efficiency and convergence rate, a simulated annealing dynamic factor is introduced to improve action selection strategies. Additionally, importance sampling is employed to evaluate different policies effectively. A case study of Shenzhen Metro is conducted to demonstrate the effectiveness of the proposed method. The results show that the method achieves convergence, fast computation speed, and real-time adjustment capabilities. Compared to traditional methods such as no adjustment, manual adjustment, and FIFO (First-In-First-Out), the proposed method significantly reduces the average total train delay by 54% and leads to more uniform train headways. The proposed method utilizes a limited number of variables for practical state descriptions, making it well suited for real-world applications. It also exhibits good scalability and transferability to other metro systems.

Validation of Artificial Intelligence (AI)-Assisted Flow Cytometry Analysis for Immunological Disorders.

Flow cytometry is a vital diagnostic tool for hematologic and immunologic disorders, but manual analysis is prone to variation and time-consuming. Over the last decade, artificial intelligence (AI) has advanced significantly. In this study, we developed and validated an AI-assisted flow cytometry workflow using 379 clinical cases from 2021, employing a 3-tube, 10-color flow panel with 21 antibodies for primary immunodeficiency diseases and related immunological disorders. The AI software (DeepFlow™, version 2.1.1) is fully automated, reducing analysis time to under 5 min per case. It interacts with hematopatholoists for manual gating adjustments when necessary. Using proprietary multidimensional density-phenotype coupling algorithm, the AI model accurately classifies and enumerates T, B, and NK cells, along with important immune cell subsets, including CD4+ helper T cells, CD8+ cytotoxic T cells, CD3+/CD4-/CD8- double-negative T cells, and class-switched or non-switched B cells. Compared to manual analysis with hematopathologist-determined lymphocyte subset percentages as the gold standard, the AI model exhibited a strong correlation (r > 0.9) across lymphocyte subsets. This study highlights the accuracy and efficiency of AI-assisted flow cytometry in diagnosing immunological disorders in a clinical setting, providing a transformative approach within a concise timeframe.

Model-free finite frequency H∞ control for active chatter suppression in turning

Regenerative chatter is an extremely harmful phenomenon in machining, which generally results in overcut and rapid tool wear. This paper presents the design, analysis, and verification of a novel model-free finite frequency band (MFFFB) H∞ control algorithm, which is dedicated to chatter suppression of turning process. The proposed MFFFB H∞ controller is established by utilizing the reinforcement Q-learning along with the generalized Kalman-Yakubovich-Popov (GKYP) lemma. Unlike available entire frequency domain (EFD) H∞ controllers, the proposed approach facilitates a less conservative implementation since the H∞ performance index is only designed for the concerned frequency band which implies the limited energies of the actuator are mainly focused on this band, and users can specify optimized frequency band according to the actual turning conditions. Furthermore, it permits the construction of H∞ controller without any knowledge of the dynamic model of turning process, only the data measured from the system state and input are required. Besides, an iterative solution algorithm for H∞ control on the basis of policy iteration is presented, which decreases the burden on manual parameter adjustment and capable of achieving the optimal solution. Simulation and experiment results demonstrate that the proposed MFFFB H∞ controller generates significant improvement in chatter-free region compared with EFD H∞ controller and conventional FFB H∞ controller.

Automatic adjustment of oxygen concentration during high-flow nasal cannula treatment using a targeted SpO2 feedback system.

To test whether targeted SpO2 feedback (TSF), an automatic control system for fraction of inspired oxygen (FiO2), achieves more time in the optimal SpO2 range and/or reduces the frequency of manual adjustments to administered FiO2 compared with conventional manual titration in patients with hypoxia on high-flow nasal cannula (HFNC) therapy. Twenty-two patients were recruited from two hospitals. For each, two sessions of manual mode and two sessions of TSF were applied in a random order, each session lasting 2 h. The target SpO2 on TSF was 95%. Oxygen monitoring levels were classified into four SpO2 ranges: hypoxia (≤ 89%), borderline (90%-93%), optimal (94%-96%) and hyperoxia (≥ 97%). The two modes were compared based on the proportion of time spent in each SpO2 range and the number of manual FiO2 adjustments. The proportion of time in the optimal SpO2 range was 20.5% under manual titration mode and 65.4% under TSF (p < .01). The proportions of time in the hypoxia range were 1.1% and 0.4%, respectively (p = .31), in the borderline range 4.7% and 3.5%, respectively (p = .54), and in the hyperoxia range 73.7% and 30.7%, respectively (p < .01). There were statistical differences only in the optimal and hyperoxia SpO2 ranges. During the 8 h, the frequency of manual FiO2 adjustment was 0.7 times for the manual mode and 0.2 times for TSF, showing no statistically significant difference (p = 0.076). Compared with manual titration, TSF achieved greater time of the optimal SpO2 and less time of hyperoxia during HFNC. The frequency of manual adjustments on TSF tended to be less than on manual titration mode. Automatic closed-loop algorithm FiO2 monitoring systems can achieve better oxygen treatments than conventional monitoring and may reduce nurse workloads. In the era of pandemic respiratory diseases, this system can also facilitate contactless SpO2 monitoring during HFNC therapy.