Preclinical evaluation of a tool for insertion force measurements in cochlear implant surgery

PURPOSE

Trauma that may be inflicted to the inner ear (cochlea) during the insertion of an electrode array (EA) in cochlear implant (CI) surgery can significantly decrease the hearing outcome of patients with residual hearing. Interaction forces between the EA and the cochlea are a promising indicator for the likelihood of intracochlear trauma. However, insertion forces have only been measured in laboratory setups. We recently developed a tool to measure the insertion force during CI surgery. Here, we present the first ex vivo evaluation of our tool with a focus on usability in the standard surgical workflow.

METHODS

Two CI surgeons inserted commercially available EAs into three temporal bone specimens. The insertion force and the orientation of the tool were recorded together with camera footage. The surgeons answered a questionnaire after each insertion to evaluate the surgical workflow with respect to CI surgery.

RESULTS

The EA insertion using our tool was rated successful in all 18 trials. The surgical workflow was evaluated to be equivalent to standard CI surgery. Minor handling challenges can be overcome through surgeon training. The peak insertion forces were 62.4 mN ± 26.7 mN on average. Peak forces significantly correlated to the final electrode insertion depth, supporting the assumption that the measured forces mainly correspond to intracochlear events and not extracochlear friction. Gravity-induced forces of up to 28.8 mN were removed from the signal, illustrating the importance of the compensation of such forces in manual surgery.

CONCLUSION

The results show that the tool is ready for intraoperative use. In vivo insertion force data will improve the interpretability of experimental results in laboratory settings. The implementation of live insertion force feedback to surgeons could further improve residual hearing preservation.

Automated Electrodermal Activity and Facial Expression Analysis for Continuous Pain Intensity Monitoring on the X-ITE Pain Database

This study focuses on improving healthcare quality by introducing an automated system that continuously monitors patient pain intensity. The system analyzes the Electrodermal Activity (EDA) sensor modality modality, compares the results obtained from both EDA and facial expressions modalities, and late fuses EDA and facial expressions modalities. This work extends our previous studies of pain intensity monitoring via an expanded analysis of the two informative methods. The EDA sensor modality and facial expression analysis play a prominent role in pain recognition; the extracted features reflect the patient's responses to different pain levels. Three different approaches were applied: Random Forest (RF) baseline methods, Long-Short Term Memory Network (LSTM), and LSTM with the sample-weighting method (LSTM-SW). Evaluation metrics included Micro average F1-score for classification and Mean Squared Error (MSE) and intraclass correlation coefficient (ICC [3, 1]) for both classification and regression. The results highlight the effectiveness of late fusion for EDA and facial expressions, particularly in almost balanced datasets (Micro average F1-score around 61%, ICC about 0.35). EDA regression models, particularly LSTM and LSTM-SW, showed superiority in imbalanced datasets and outperformed guessing (where the majority of votes indicate no pain) and baseline methods (RF indicates Random Forest classifier (RFc) and Random Forest regression (RFr)). In conclusion, by integrating both modalities or utilizing EDA, they can provide medical centers with reliable and valuable insights into patients' pain experiences and responses.

On the Stability and Homogeneous Ensemble of Feature Selection for Predictive Maintenance: A Classification Application for Tool Condition Monitoring in Milling

Feature selection (FS) represents an essential step for many machine learning-based predictive maintenance (PdM) applications, including various industrial processes, components, and monitoring tasks. The selected features not only serve as inputs to the learning models but also can influence further decisions and analysis, e.g., sensor selection and understandability of the PdM system. Hence, before deploying the PdM system, it is crucial to examine the reproducibility and robustness of the selected features under variations in the input data. This is particularly critical for real-world datasets with a low sample-to-dimension ratio (SDR). However, to the best of our knowledge, stability of the FS methods under data variations has not been considered yet in the field of PdM. This paper addresses this issue with an application to tool condition monitoring in milling, where classifiers based on support vector machines and random forest were employed. We used a five-fold cross-validation to evaluate three popular filter-based FS methods, namely Fisher score, minimum redundancy maximum relevance (mRMR), and ReliefF, in terms of both stability and macro-F1. Further, for each method, we investigated the impact of the homogeneous FS ensemble on both performance indicators. To gain broad insights, we used four (2:2) milling datasets obtained from our experiments and NASA's repository, which differ in the operating conditions, sensors, SDR, number of classes, etc. For each dataset, the study was conducted for two individual sensors and their fusion. Among the conclusions: (1) Different FS methods can yield comparable macro-F1 yet considerably different FS stability values. (2) Fisher score (single and/or ensemble) is superior in most of the cases. (3) mRMR's stability is overall the lowest, the most variable over different settings (e.g., sensor(s), subset cardinality), and the one that benefits the most from the ensemble.

AIMSurv: First pan-European harmonized surveillance of Aedes invasive mosquito species of relevance for human vector-borne diseases

Human and animal vector-borne diseases, particularly mosquito-borne diseases, are emerging or re-emerging worldwide. Six Aedes invasive mosquito (AIM) species were introduced to Europe since the 1970s: Aedes aegypti, Ae. albopictus, Ae. japonicus, Ae. koreicus, Ae. atropalpus and Ae. triseriatus. Here, we report the results of AIMSurv2020, the first pan-European surveillance effort for AIMs. Implemented by 42 volunteer teams from 24 countries. And presented in the form of a dataset named "AIMSurv Aedes Invasive Mosquito species harmonized surveillance in Europe. AIM-COST Action. Project ID: CA17108". AIMSurv2020 harmonizes field surveillance methodologies for sampling different AIMs life stages, frequency and minimum length of sampling period, and data reporting. Data include minimum requirements for sample types and recommended requirements for those teams with more resources. Data are published as a Darwin Core archive in the Global Biodiversity Information Facility- Spain, comprising a core file with 19,130 records (EventID) and an occurrences file with 19,743 records (OccurrenceID). AIM species recorded in AIMSurv2020 were Ae. albopictus, Ae. japonicus and Ae. koreicus, as well as native mosquito species.

How to predict relapse in leukemia using time series data: A comparative in silico study

Risk stratification and treatment decisions for leukemia patients are regularly based on clinical markers determined at diagnosis, while measurements on system dynamics are often neglected. However, there is increasing evidence that linking quantitative time-course information to disease outcomes can improve the predictions for patient-specific treatment responses. We designed a synthetic experiment simulating response kinetics of 5,000 patients to compare different computational methods with respect to their ability to accurately predict relapse for chronic and acute myeloid leukemia treatment. Technically, we used clinical reference data to first fit a model and then generate de novo model simulations of individual patients' time courses for which we can systematically tune data quality (i.e. measurement error) and quantity (i.e. number of measurements). Based hereon, we compared the prediction accuracy of three different computational methods, namely mechanistic models, generalized linear models, and deep neural networks that have been fitted to the reference data. Reaching prediction accuracies between 60 and close to 100%, our results indicate that data quality has a higher impact on prediction accuracy than the specific choice of the particular method. We further show that adapted treatment and measurement schemes can considerably improve the prediction accuracy by 10 to 20%. Our proof-of-principle study highlights how computational methods and optimized data acquisition strategies can improve risk assessment and treatment of leukemia patients.

Knowledge on exotic mosquitoes in Germany, and public acceptance and effectiveness of Bti and two self-prepared insecticides against Aedes japonicus japonicus

Mosquito-borne diseases are a continuous challenge to public health. To prevent transmission, Integrated Vector Management (IVM) applies preventive, control, and communicational strategies that should be feasible, environmentally benign, and sustainable. IVM shows higher efficiency when being supported by local communities. Accordingly, we applied a social-ecological approach to identify the public acceptance of control measures and effectiveness of Eurocent coins containing copper, clove essential oil (EO) and Bacillus thuringiensis israelensis (Bti). We performed field and laboratory experiments to demonstrate the toxicity of alternative substances against Aedes japonicus japonicus. In expert interviews, we asked for (1) knowledge on exotic mosquitoes in Germany, (2) potential chances of alternative substances in future mosquito control, and (3) their needs for further clarification before application. We assessed potential users' (4) awareness of exotic mosquitoes and (5) willingness to apply the substances. Self-prepared copper coins and EO were clearly preferred by potential users over Bti. However, 100% mortality of the sensitive first stage could not be reached with the number of ten 5-Eurocent coins showing limited toxicity. Clove EO was shown to work as oviposition deterrent and larvicide with a LC50 of 17 mg l-1 (95% CI: 15-19 mg l-1). This study shows the importance of potential users' perspectives in IVM and the need for authorised insecticides.

Development and Characterization of Defect-Free Matrimid® Mixed-Matrix Membranes Containing Activated Carbon Particles for Gas Separation

In this work, mixed-matrix membranes (MMMs) for gas separation in the form of thick films were prepared via the combination of the polymer Matrimid® 5218 and activated carbons (AC). The AC particles had a mean particle size of 1.5 μm and a mean pore diameter of 1.9 nm. The films were prepared by slow solvent evaporation from casting solutions in chloroform, which had a varying polymer⁻AC ratio. It was possible to produce stable films with up to a content of 50 vol % of AC. Thorough characterization experiments were accomplished via differential scanning calorimetry and thermogravimetric analysis, while the morphology of the MMMs was also investigated via scanning electron microscopy. The gas transport properties were revealed by employing time-lag measurements for different pure gases as well as sorption balance experiments for the filler particles. It was found that defect free Matrimid® MMMs with AC were prepared and the increase of the filler content led to a higher effective permeability for different gases. The single gas selectivity αij of different gas pairs maintained stable values with the increase of AC content, regardless of the steep increase in the effective permeability of the pure gases. Estimation of the solubilities and the diffusivities of the Matrimid®, AC, and MMMs allowed for the explanation of the increasing permeabilities of the MMMs, with the increase of AC content by modelling.