Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

A general force field by machine learning on experimental crystal structures. Calculations of intermolecular Gibbs energy with FlexCryst.

In Acta crystallographica. Section A, Foundations and advances

Machine learning was employed on the experimental crystal structures of the Cambridge Structural Database (CSD) to derive an intermolecular force field for all available types of atoms (general force field). The obtained pairwise interatomic potentials of the general force field allow for the fast and accurate calculation of intermolecular Gibbs energy. The approach is based on three postulates regarding Gibbs energy: the lattice energy must be below zero, the crystal structure must be a local minimum, and, if available, the experimental and the calculated lattice energy must coincide. The parametrized general force field was then validated regarding these three conditions. First, the experimental lattice energy was compared with the calculated energies. The observed errors were found to be in the order of experimental errors. Second, Gibbs lattice energy was calculated for all structures available in the CSD. Their energy values were found to be below zero in 99.86% of the cases. Finally, 500 random structures were minimized, and the change in density and energy was examined. The mean error in the case of density was below 4.06%, and for energy it was below 5.7%. The obtained general force field calculated Gibbs lattice energies of 259 041 known crystal structures within a few hours. Since Gibbs energy defines the reaction energy, the calculated energy can be used to predict chemical-physical properties of crystals, for instance, the formation of co-crystals, polymorph stability and solubility.

Hofmann Detlef Walter Maria, Kuleshova Liudmila Nikolaevna

2023-Mar-01

Gibbs energy, crystal structure analysis, force field, intermolecular interactions, machine learning

General General

Use of Bayesian decision analysis to maximize value in patient-centered randomized clinical trials in Parkinson's disease.

In Journal of biopharmaceutical statistics

A fixed one-sided significance level of 5% is commonly used to interpret the statistical significance of randomized clinical trial (RCT) outcomes. While it is necessary to reduce the false positive rate, the threshold used could be chosen quantitatively and transparently to specifically reflect patient preferences regarding benefit-risk tradeoffs as well as other considerations. How can patient preferences be explicitly incorporated into RCTs in Parkinson's disease (PD), and what is the impact on statistical thresholds for device approval? In this analysis, we apply Bayesian decision analysis (BDA) to PD patient preference scores elicited from survey data. BDA allows us to choose a sample size (n) and significance level (α) that maximizes the overall expected value to patients of a balanced two-arm fixed-sample RCT, where the expected value is computed under both null and alternative hypotheses. For PD patients who had previously received deep brain stimulation (DBS) treatment, the BDA-optimal significance levels fell between 4.0% and 10.0%, similar to or greater than the traditional value of 5%. Conversely, for patients who had never received DBS, the optimal significance level ranged from 0.2% to 4.4%. In both of these populations, the optimal significance level increased with the severity of the patients' cognitive and motor function symptoms. By explicitly incorporating patient preferences into clinical trial designs and the regulatory decision-making process, BDA provides a quantitative and transparent approach to combine clinical and statistical significance. For PD patients who have never received DBS treatment, a 5% significance threshold may not be conservative enough to reflect their risk-aversion level. However, this study shows that patients who previously received DBS treatment present a higher tolerance to accept therapeutic risks in exchange for improved efficacy which is reflected in a higher statistical threshold.

Chaudhuri Shomesh E, Ben Chaouch Zied, Hauber Brett, Mange Brennan, Zhou Mo, Christopher Stephanie, Bardot Dawn, Sheehan Margaret, Donnelly Anne, McLaughlin Lauren, Caldwell Brittany, Benz Heather L, Ho Martin, Saha Anindita, Gwinn Katrina, Sheldon Murray, Lo Andrew W

2023-Mar-02

Bayesian decision analysis, Parkinson’s disease, benefit–risk, medical devices, patient preferences

Radiology Radiology

Application of logistic regression and machine learning methods for idiopathic inflammatory myopathies malignancy prediction.

In Clinical and experimental rheumatology ; h5-index 43.0

OBJECTIVES : Malignancy is related to idiopathic inflammatory myopathies (IIM) and leads to a poor prognosis. Early prediction of malignancy is thought to improve the prognosis. However, predictive models have rarely been reported in IIM. Herein, we aimed to establish and use a machine learning (ML) algorithm to predict the possible risk factors for malignancy in IIM patients.

METHODS : We retrospectively reviewed the medical records of 168 patients diagnosed with IIM in Shantou Central hospital, from 2013 to 2021. We randomly divided patients into two groups, the training sets (70%) for construction of the prediction model, and the validation sets (30%) for evaluation of model performance. We constructed six types of ML algorithms models and the AUC of ROC curves were used to describe the efficacy of the model. Finally, we set up a web version using the best prediction model to make it more generally available.

RESULTS : According to the multi-variable regression analysis, three predictors were found to be the risk factors to establish the prediction model, including age, ALT<80U/L, and anti-TIF1-γ, and ILD was found to be a protective factor. Compared with five other ML algorithms models, the traditional algorithm logistic regression (LR) model was as good or better than the other models to predict malignancy in IIM. The AUC of the ROC using LR was 0.900 in the training set and 0.784 in the validation set. We selected the LR model as the final prediction model. Accordingly, a nomogram was constructed using the above four factors. A web version was built and can be visited on the website or acquired by scanning the QR code.

CONCLUSIONS : The LR algorithm appears to be a good predictor of malignancy and may help clinicians screen, evaluate and follow up high-risk patients with IIM.

Zhang Weijin, Huang Guohai, Zheng Kedi, Lin Jianqun, Hu Shijian, Zheng Shaoyu, Du Guangzhou, Zhang Guohong, Bruni Cosimo, Matucci-Cerinic Marco, Furst Daniel E, Wang Yukai

2023-Mar

General General

Freeform optical system design with differentiable three-dimensional ray tracing and unsupervised learning.

In Optics express

Optical systems have been crucial for versatile applications such as consumer electronics, remote sensing and biomedical imaging. Designing optical systems has been a highly professional work due to complicated aberration theories and intangible rules-of-thumb, hence neural networks are only coming into this realm until recent years. In this work, we propose and implement a generic, differentiable freeform raytracing module, suitable for off-axis, multiple-surface freeform/aspheric optical systems, paving the way toward a deep learning-based optical design method. The network is trained with minimal prior knowledge, and it can infer numerous optical systems after a one-time training. The presented work unlocks great potential for deep learning in various freeform/aspheric optical systems, and the trained network could serve as an effective, unified platform for generating, recording, and replicating good initial optical designs.

Nie Yunfeng, Zhang Jingang, Su Runmu, Ottevaere Heidi

2023-Feb-27

Radiology Radiology

Robust framework for COVID-19 identication from a multicenter dataset of chest CT scans.

In PloS one ; h5-index 176.0

The main objective of this study is to develop a robust deep learning-based framework to distinguish COVID-19, Community-Acquired Pneumonia (CAP), and Normal cases based on volumetric chest CT scans, which are acquired in different imaging centers using different scanners and technical settings. We demonstrated that while our proposed model is trained on a relatively small dataset acquired from only one imaging center using a specific scanning protocol, it performs well on heterogeneous test sets obtained by multiple scanners using different technical parameters. We also showed that the model can be updated via an unsupervised approach to cope with the data shift between the train and test sets and enhance the robustness of the model upon receiving a new external dataset from a different center. More specifically, we extracted the subset of the test images for which the model generated a confident prediction and used the extracted subset along with the training set to retrain and update the benchmark model (the model trained on the initial train set). Finally, we adopted an ensemble architecture to aggregate the predictions from multiple versions of the model. For initial training and development purposes, an in-house dataset of 171 COVID-19, 60 CAP, and 76 Normal cases was used, which contained volumetric CT scans acquired from one imaging center using a single scanning protocol and standard radiation dose. To evaluate the model, we collected four different test sets retrospectively to investigate the effects of the shifts in the data characteristics on the model's performance. Among the test cases, there were CT scans with similar characteristics as the train set as well as noisy low-dose and ultra-low-dose CT scans. In addition, some test CT scans were obtained from patients with a history of cardiovascular diseases or surgeries. This dataset is referred to as the "SPGC-COVID" dataset. The entire test dataset used in this study contains 51 COVID-19, 28 CAP, and 51 Normal cases. Experimental results indicate that our proposed framework performs well on all test sets achieving total accuracy of 96.15% (95%CI: [91.25-98.74]), COVID-19 sensitivity of 96.08% (95%CI: [86.54-99.5]), CAP sensitivity of 92.86% (95%CI: [76.50-99.19]), Normal sensitivity of 98.04% (95%CI: [89.55-99.95]) while the confidence intervals are obtained using the significance level of 0.05. The obtained AUC values (One class vs Others) are 0.993 (95%CI: [0.977-1]), 0.989 (95%CI: [0.962-1]), and 0.990 (95%CI: [0.971-1]) for COVID-19, CAP, and Normal classes, respectively. The experimental results also demonstrate the capability of the proposed unsupervised enhancement approach in improving the performance and robustness of the model when being evaluated on varied external test sets.

Khademi Sadaf, Heidarian Shahin, Afshar Parnian, Enshaei Nastaran, Naderkhani Farnoosh, Rafiee Moezedin Javad, Oikonomou Anastasia, Shafiee Akbar, Babaki Fard Faranak, Plataniotis Konstantinos N, Mohammadi Arash

2023

Surgery Surgery

Spatiotemporal modeling of grip forces captures proficiency in manual robot control

Bioengineering, 2023; 10(1):59

This paper builds on our previous work by exploiting Artificial Intelligence to predict individual grip force variability in manual robot control. Grip forces were recorded from various loci in the dominant and non dominant hands of individuals by means of wearable wireless sensor technology. Statistical analyses bring to the fore skill specific temporal variations in thousands of grip forces of a complete novice and a highly proficient expert in manual robot control. A brain inspired neural network model that uses the output metric of a Self Organizing Map with unsupervised winner take all learning was run on the sensor output from both hands of each user. The neural network metric expresses the difference between an input representation and its model representation at any given moment in time t and reliably captures the differences between novice and expert performance in terms of grip force variability.Functionally motivated spatiotemporal analysis of individual average grip forces, computed for time windows of constant size in the output of a restricted amount of task-relevant sensors in the dominant (preferred) hand, reveal finger-specific synergies reflecting robotic task skill. The analyses lead the way towards grip force monitoring in real time to permit tracking task skill evolution in trainees, or identify individual proficiency levels in human robot interaction in environmental contexts of high sensory uncertainty. Parsimonious Artificial Intelligence (AI) assistance will contribute to the outcome of new types of surgery, in particular single-port approaches such as NOTES (Natural Orifice Transluminal Endoscopic Surgery) and SILS (Single Incision Laparoscopic Surgery).

Rongrong Liu, John M. Wandeto, Florent Nageotte, Philippe Zanne, Michel de Mathelin, Birgitta Dresp-Langley

2023-03-03