Traction force profile in children with severe perinatal outcomes delivered with a digital vacuum extraction handle: A case-control study.

INTRODUCTION: During the second stage of labor, vacuum-assisted delivery is an alternative to forceps delivery and emergency cesarean section. Extensive research concerning perinatal outcomes has indicated that the risk of complications, although rare, is higher than with a spontaneous vaginal delivery. An important factor related to perinatal outcomes is the traction force applied. Our research group previously developed a digital extraction handle, the Vacuum Intelligent Handle-3 (VIH3), that measures and records traction force. The objective of this study was to compare traction force profiles in children with and without severe perinatal outcomes delivered with the digital handle. A secondary aim was to establish a safe force limit. MATERIAL AND METHODS: This was an observational case-control study at the delivery ward at Karolinska University Hospital, Sweden. In total, 573 children delivered with the digital handle between 2012 and 2018 were included. Cases were defined as a composite of severe perinatal outcomes, including subgaleal hematoma, intracranial hemorrhage, hypoxic ischemic encephalopathy 1-3, seizures or death. The cases in the cohort were matched 1:3 based on five matching variables. Traction profiles were analyzed using the MATLAB® software and conditional logistic regression. RESULTS: The incidence of severe perinatal outcomes was 2.3%. The 13 cases were matched with three controls each (n = 39). A statistically significant increased odds for higher total traction forces was seen in the case group (odds ratio [OR] 1.004; 95% confidence interval [CI] 1.001-1.007) and for the peak force (OR 1.022; 95% CI 1.004-1.041). Several procedure-related parameters were significantly increased in the case group. As expected, some neonatal characteristics also differed significantly. An upper force limit of 343 Newton minutes (Nmin) revealed an 86% reduction in severe perinatal outcomes (adjusted OR 0.14; 95% CI 0.04-0.5). CONCLUSIONS: Children with severe perinatal outcomes had traction force profiles with significantly higher forces. The odds for severe perinatal outcomes increased for every increase in Nmin and Newton used during the extraction procedure. A calculated total force level of 343 Nmin is suggested as an upper safety limit, but this must be tested prospectively to provide validity.

Perinatal outcome after vacuum assisted delivery with digital feedback on traction force; a randomised controlled study.

BACKGROUND: Low and mid station vacuum assisted deliveries (VAD) are delicate manual procedures that entail a high degree of subjectivity from the operator and are associated with adverse neonatal outcome. Little has been done to improve the procedure, including the technical development, traction force and the possibility of objective documentation. We aimed to explore if a digital handle with instant haptic feedback on traction force would reduce the neonatal risk during low or mid station VAD. METHODS: A two centre, randomised superiority trial at Karolinska University Hospital, Sweden, 2016-2018. Cases were randomised bedside to either a conventional or a digital handle attached to a Bird metal cup (50 mm, 80 kPa). The digital handle measured applied force including an instant notification by vibration when high levels of traction force were predicted according to a predefined algorithm. Primary outcome was a composite of hypoxic ischaemic encephalopathy, intracranial haemorrhage, seizures, death and/or subgaleal hematoma. Three hundred eighty low and mid VAD in each group were estimated to decrease primary outcome from six to 2 %. RESULTS: After 2 years, an interim analyse was undertaken. Meeting the inclusion criteria, 567 vacuum extractions were randomized to the use of a digital handle (n = 296) or a conventional handle (n = 271). Primary outcome did not differ between the two groups: (2.7% digital handle vs 2.6% conventional handle). The incidence of primary outcome differed significantly between the two delivery wards (4% vs 0.9%, p < 0.05). A recalculation of power revealed that 800 cases would be needed in each group to show a decrease in primary outcome from three to 1 %. This was not feasible, and the study therefore closed. CONCLUSIONS: The incidence of primary outcome was lower than estimated and the study was underpowered. However, the difference between the two delivery wards might reflect varying degree of experience of the technical equipment. An objective documentation of the extraction procedure is an attractive alternative in respect to safety and clinical training. To demonstrate improved safety, a multicentre study is required to reach an adequate cohort. This was beyond the scope of the study. TRIAL REGISTRATION: ClinicalTrials.gov NCT03071783 , March 1, 2017, retrospectively registered.

An enhancement algorithm for head characteristics of caged chickens detection based on cyclic consistent migration neural network.

The enclosed multistory poultry housing is a type of poultry enclosure widely used in industrial caged chicken breeding. Accurate identification and detection of the comb and eyes of caged chickens in poultry farms using this type of enclosure can enhance managers' understanding of the health of caged chickens. However, the accuracy of image detection of caged chickens will be affected by the enclosure's entrance, which will reduce the precision. Therefore, this paper proposes a cage-gate removal algorithm based on big data and deep learning Cyclic Consistent Migration Neural Network (CCMNN). The method achieves automatic elimination and restoration of some key information in the image through the CCMNN network. The Structural Similarity Index Measure (SSIM) between the recovered and original images on the test set is 91.14%. Peak signal-to-noise ratio (PSNR) is 25.34dB. To verify the practicability of the proposed method, the performance of the target detection algorithm is analyzed both before and after applying the CCMNN network in detecting the combs and eyes of caged chickens. Different YOLOv8 detection algorithms, including YOLOv8s, YOLOv8n, YOLOv8m, and YOLOv8x, were used to verify the algorithm proposed in this paper. The experimental results demonstrate that compared to images without CCMNN processing, the precision of comb detection of caged chickens is improved by 11, 11.3, 12.8, and 10.2%. Similarly, the precision of eye detection for caged chickens is improved by 2.4, 10.2, 6.8, and 9%. Therefore, more complete outline images of caged chickens can be obtained using this algorithm and the precision in detecting the comb and eyes of caged chickens can be enhanced. These advancements in the algorithm offer valuable insights for future poultry researchers aiming to deploy enhanced detection equipment, thereby contributing to the accurate assessment of poultry production and farm conditions.

On-line weight estimation of broiler carcass and cuts by a computer vision system.

In a broiler carcass production conveyor system, inspection, monitoring, and grading carcass and cuts based on computer vision techniques are challenging due to cuts segmentation and ambient light conditions issues. This study presents a depth image-based broiler carcass weight prediction system. An Active Shape Model was developed to segment the carcass into 4 cuts (drumsticks, breasts, wings, and head and neck). Five regression models were developed based on the image features for each weight estimation (carcass and its cuts). The Bayesian-ANN model outperformed all other regression models at 0.9981 R2 and 0.9847 R2 in the whole carcass and head and neck weight estimation. The RBF-SVR model surpassed all the other drumstick, breast, and wings weight prediction models at 0.9129 R2, 0.9352 R2, and 0.9896 R2, respectively. This proposed technique can be applied as a nondestructive, nonintrusive, and accurate on-line broiler carcass production system in the automation of chicken carcass and cuts weight estimation.

Efficient pyrolysis of ginkgo biloba leaf residue and pharmaceutical sludge (mixture) with high production of clean energy: Process optimization by particle swarm optimization and gradient boosting decision tree algorithm.

Production of sustainable clean energy can be achieved by co-pyrolysis of agricultural residues and wastewater sludge. Herein, non-additive thermal behaviour of co-pyrolysis of pharmaceutical sludge and ginkgo biloba leaf residues was investigated. Synergistic effect of co-pyrolysis was not obvious at elevated temperatures. Further, kinetics of co-pyrolysis was studied by fitting Coats-Redfern integration method to thermogravimetric (TG) curve. The change of heat and mass transfer in the reactor caused the change of dynamic parameters. Moreover, hybrid particle swarm optimization and gradient boosting decision tree (PSO-GBDT) algorithm was designed to boost the energy production at full-scale pyrolysis plant by monitoring TG curves. PSO-GBDT model well predicts mass loss rate of the mixture at different heating rates confirming that co-pyrolysis of PS and GBLR can results in high energy production by increasing PS pyrolysis. Designing PSO-GBDT model help to reduced waste production by resourceful treatment of waste in to energy.

Broiler stunned state detection based on an improved fast region-based convolutional neural network algorithm.

An improved fast region-based convolutional neural network (RCNN) algorithm is proposed to improve the accuracy and efficiency of recognizing broilers in a stunned state. The algorithm recognizes 3 stunned state conditions: insufficiently stunned, moderately stunned, and excessively stunned. Image samples of stunned broilers were collected from a slaughter line using an image acquisition platform. According to the format of PASCAL VOC (pattern analysis, statistical modeling, and computational learning visual object classes) dataset, a dataset for each broiler stunned state condition was obtained using an annotation tool to mark the chicken head and wing area in the original image. A rotation and flip data augmentation method was used to enhance the effectiveness of the datasets. Based on the principle of a residual network, a multi-layer residual module (MRM) was constructed to facilitate more detailed feature extraction. A model was then developed (entitled here Faster-RCNN+MRMnet) and used to detect broiler stunned state conditions. When applied to a reinforcing dataset containing 27,828 images of chickens in a stunned state, the identification accuracy of the model was 98.06%. This was significantly higher than both the established back propagation neural network model (90.11%) and another Faster-RCNN model (96.86%). The proposed algorithm can complete the inspection of the stunned state of more than 40,000 broilers per hour. The approach can be used for online inspection applications to increase efficiency, reduce labor and cost, and yield significant benefits for poultry processing plants.

Predictive value of traction force measurement in vacuum extraction: Development of a multivariate prognostic model.

OBJECTIVE: To enable early prediction of strong traction force vacuum extraction. DESIGN: Observational cohort. SETTING: Karolinska University Hospital delivery ward, tertiary unit. POPULATION AND SAMPLE SIZE: Term mid and low metal cup vacuum extraction deliveries June 2012-February 2015, n = 277. METHODS: Traction forces during vacuum extraction were collected prospectively using an intelligent handle. Levels of traction force were analysed pairwise by subjective category strong versus non-strong extraction, in order to define an objective predictive value for strong extraction. STATISTICAL ANALYSIS: A logistic regression model based on the shrinkage and selection method lasso was used to identify the predictive capacity of the different traction force variables. PREDICTORS: Total (time force integral, Newton minutes) and peak traction (Newton) force in the first to third pull; difference in traction force between the second and first pull, as well as the third and first pull respectively. Accumulated traction force at the second and third pull. OUTCOME: Subjectively categorized extraction as strong versus non-strong. RESULTS: The prevalence of strong extraction was 26%. Prediction including the first and second pull: AUC 0,85 (CI 0,80-0,90); specificity 0,76; sensitivity 0,87; PPV 0,56; NPV 0,94. Prediction including the first to third pull: AUC 0,86 (CI 0,80-0,91); specificity 0,87; sensitivity 0,70; PPV 0,65; NPV 0,89. CONCLUSION: Traction force measurement during vacuum extraction can help exclude strong category extraction from the second pull. From the third pull, two-thirds of strong extractions can be predicted.