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

Radiology Radiology

Mortality prediction for patients with acute respiratory distress syndrome based on machine learning: a population-based study.

In Annals of translational medicine

Background : Traditional scoring systems for patients' outcome prediction in intensive care units such as Oxygenation Saturation Index (OSI) and Oxygenation Index (OI) may not reliably predict the clinical prognosis of patients with acute respiratory distress syndrome (ARDS). Thus, none of them have been widely accepted for mortality prediction in ARDS. This study aimed to develop and validate a mortality prediction method for patients with ARDS based on machine learning using the Medical Information Mart for Intensive Care (MIMIC-III) and Telehealth Intensive Care Unit (eICU) Collaborative Research Database (eICU-CRD) databases.

Methods : Patients with ARDS were selected based on the Berlin definition in MIMIC-III and eICU-CRD databases. The APPS score (using age, PaO2/FiO2, and plateau pressure), Simplified Acute Physiology Score II (SAPS-II), Sepsis-related Organ Failure Assessment (SOFA), OSI, and OI were calculated. With MIMIC-III data, a mortality prediction model was built based on the random forest (RF) algorithm, and the performance was compared to those of existing scoring systems based on logistic regression. The performance of the proposed RF method was also validated with the combined MIMIC-III and eICU-CRD data. The performance of mortality prediction was evaluated by using the area under the receiver operating characteristics curve (AUROC) and performing calibration using the Hosmer-Lemeshow test.

Results : With the MIMIC-III dataset (308 patients, for comparisons with the existing scoring systems), the RF model predicted the in-hospital mortality, 30-day mortality, and 1-year mortality with an AUROC of 0.891, 0.883, and 0.892, respectively, which were significantly higher than those of the SAPS-II, APPS, OSI, and OI (all P<0.001). In the multi-source validation (the combined dataset of 2,235 patients in MIMIC-III and 331 patients in eICU-CRD), the RF model achieved an AUROC of 0.905 and 0.736 for predicting in-hospital mortality for the MIMIC-III and eICU-CRD datasets, respectively. The calibration plots suggested good fits for our RF model and these scoring systems for predicting mortality. The platelet count and lactate level were the strongest predictive variables for predicting in-hospital mortality.

Conclusions : Compared to the existing scoring systems, machine learning significantly improved performance for predicting ARDS mortality. Validation with multi-source datasets showed a relatively robust generalisation ability of our prediction model.

Huang Bingsheng, Liang Dong, Zou Rushi, Yu Xiaxia, Dan Guo, Huang Haofan, Liu Heng, Liu Yong


Acute respiratory distress syndrome (ARDS), intensive care unit (ICU), machine learning (ML), mortality

Surgery Surgery

Probabilistic ratiocination of hepatocellular carcinoma after resection: evaluation of expected to be promising approaches.

In Annals of translational medicine

Background : Precise prediction of survival after treatment is of great importance for patients with diseases with high mortality. RNA sequencing data and deep learning (DL) methods are expected to become promising approaches in the development of prediction models in the future. We aimed to evaluate the optimal covariates and methodology for patients with hepatocellular carcinoma (HCC) undergoing surgical resection.

Methods : The Cox proportional hazards regression model and the DL approach were used to develop prediction models incorporating clinical, genetic, and combined clinical and genetic variables for survival prediction in patients with HCC after resection. A total of 1,114 patients and 184 patients were enrolled in the present study from 2,163 and 601 patients from Eastern Hepatobiliary Surgery Hospital and Renji Hospital, respectively. The models were internally validated through random sampling and externally validated in clinical cohorts. Between-model comparisons were carried out in terms of the integrated discrimination improvement and net reclassification index.

Results : The Cox and DL clinical models were developed by adopting 7 independent prognostic factors (total bilirubin, prothrombin time, tumor size, tumor number, lymph node metastasis, and vascular invasion) and 22 clinical factors, respectively. Both the Cox clinical model and the DL clinical model showed excellent performances in the derivation [area under the curve (AUC): 0.75 vs. 0.77] and validation (AUC: 0.83 vs. 0.80) sets. The derived Cox genetic model with 6 significant prognostic genes was not as effective as the DL approach involving 686 genes. A combined clinical and genetic approach modified the performances of both the Cox and DL models. The integrated discrimination improvement and net reclassification index of the DL clinical model were generally better than those of the Cox clinical model.

Conclusions : Our Cox clinical model sufficiently provided precise survival prediction in patients with HCC after resection. It may serve as an accurate and cost-effective tool for predicting survival in such patients.

Dong Wei, Guo Xinggang, Liu Fuchen, Zhang Wenli, Wang Zongyan, Tian Tao, Tao Qifei, Hou Guojun, Zhou Weiping, Jeong Seogsong, Xia Qiang, Liu Hui


Hepatocellular carcinoma (HCC), nomogram, predictive systems, surgical resection, survival outcomes

General General

Artificial intelligence in digital cariology: a new tool for the diagnosis of deep caries and pulpitis using convolutional neural networks.

In Annals of translational medicine

Background : An accurate diagnosis of deep caries and pulpitis on periapical radiographs is a clinical challenge.

Methods : A total of 844 radiographs were included in this study. Of the 844, 717 (85%) were used for training and 127 (15%) were used for testing the three convolutional neural networks (CNNs) (VGG19, Inception V3, and ResNet18). The performance [accuracy, precision, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC)] of the CNNs were evaluated and compared. The CNN model with the best performance was further integrated with clinical parameters to see whether multi-modal CNN could provide an enhanced performance. The Gradient-weighted Class Activation Mapping (Grad-CAM) technique illustrates what image feature was the most important for the CNNs.

Results : The CNN of ResNet18 demonstrated the best performance [accuracy =0.82, 95% confidence interval (CI): 0.80-0.84; precision =0.81, 95% CI: 0.73-0.89; sensitivity =0.85, 95% CI: 0.79-0.91; specificity =0.82, 95% CI: 0.76-0.88; and AUC =0.89, 95% CI: 0.86-0.92], compared with VGG19 and Inception V3 as well as the comparator dentists. Therefore, ResNet18 was chosen to integrate with clinical parameters to produce the multi-modal CNN of ResNet18 + C, which showed a significantly enhanced performance (accuracy =0.86, 95% CI: 0.84-0.88; precision =0.85, 95% CI: 0.76-0.94; sensitivity =0.89, 95% CI: 0.83-0.95; specificity =0.86, 95% CI: 0.79-0.93; and AUC =0.94, 95% CI: 0.91-0.97).

Conclusions : The CNN of ResNet18 showed good performance (accuracy, precision, sensitivity, specificity, and AUC) for the diagnosis of deep caries and pulpitis. The multi-modal CNN of ResNet18 + C (ResNet18 integrated with clinical parameters) demonstrated a significantly enhanced performance, with promising potential for the diagnosis of deep caries and pulpitis.

Zheng Liwen, Wang Haolin, Mei Li, Chen Qiuman, Zhang Yuxin, Zhang Hongmei


Artificial intelligence (AI), caries, carious lesions, convolutional neural network (CNNs), deep learning, pulpitis

General General

Using Tactile Sensing to Improve the Sample Efficiency and Performance of Deep Deterministic Policy Gradients for Simulated In-Hand Manipulation Tasks.

In Frontiers in robotics and AI

Deep Reinforcement Learning techniques demonstrate advances in the domain of robotics. One of the limiting factors is a large number of interaction samples usually required for training in simulated and real-world environments. In this work, we demonstrate for a set of simulated dexterous in-hand object manipulation tasks that tactile information can substantially increase sample efficiency for training (by up to more than threefold). We also observe an improvement in performance (up to 46%) after adding tactile information. To examine the role of tactile-sensor parameters in these improvements, we included experiments with varied sensor-measurement accuracy (ground truth continuous values, noisy continuous values, Boolean values), and varied spatial resolution of the tactile sensors (927 sensors, 92 sensors, and 16 pooled sensor areas in the hand). To facilitate further studies and comparisons, we make these touch-sensor extensions available as a part of the OpenAI Gym Shadow-Dexterous-Hand robotics environments.

Melnik Andrew, Lach Luca, Plappert Matthias, Korthals Timo, Haschke Robert, Ritter Helge


deep learning, in-hand manipulation, reinforcement learning, robotics, sample-efficiency, shadow dexterous hand, tactile sensing

General General

Multicolor image classification using the multimodal information bottleneck network (MMIB-Net) for detecting diabetic retinopathy.

In Optics express

Multicolor (MC) imaging is an imaging modality that records confocal scanning laser ophthalmoscope (cSLO) fundus images, which can be used for the diabetic retinopathy (DR) detection. By utilizing this imaging technique, multiple modal images can be obtained in a single case. Additional symptomatic features can be obtained if these images are considered during the diagnosis of DR. However, few studies have been carried out to classify MC Images using deep learning methods, let alone using multi modal features for analysis. In this work, we propose a novel model which uses the multimodal information bottleneck network (MMIB-Net) to classify the MC Images for the detection of DR. Our model can extract the features of multiple modalities simultaneously while finding concise feature representations of each modality using the information bottleneck theory. MC Images classification can be achieved by picking up the combined representations and features of all modalities. In our experiments, it is shown that the proposed method can achieve an accurate classification of MC Images. Comparative experiments also demonstrate that the use of multimodality and information bottleneck improves the performance of MC Images classification. To the best of our knowledge, this is the first report of DR identification utilizing the multimodal information bottleneck convolutional neural network in MC Images.

Song Jingqi, Zheng Yuanjie, Wang Jing, Zakir Ullah Muhammad, Jiao Wanzhen


Pathology Pathology

Computational pathology reveals unique spatial patterns of immune response in H&E images from COVID-19 autopsies: preliminary findings.

In Journal of medical imaging (Bellingham, Wash.)

Purpose: We used computerized image analysis and machine learning approaches to characterize spatial arrangement features of the immune response from digitized autopsied H&E tissue images of the lung in coronavirus disease 2019 (COVID-19) patients. Additionally, we applied our approach to tease out potential morphometric differences from autopsies of patients who succumbed to COVID-19 versus H1N1. Approach: H&E lung whole slide images from autopsy specimens of nine COVID-19 and two H1N1 patients were computationally interrogated. 606 image patches ( 55 per patient) of 1024 × 882    pixels were extracted from the 11 autopsied patient studies. A watershed-based segmentation approach in conjunction with a machine learning classifier was employed to identify two types of nuclei families: lymphocytes and non-lymphocytes (i.e., other nucleated cells such as pneumocytes, macrophages, and neutrophils). Based off the proximity of the individual nuclei, clusters for each nuclei family were constructed. For each of the resulting clusters, a series of quantitative measurements relating to architecture and density of nuclei clusters were calculated. A receiver operating characteristics-based feature selection method, violin plots, and the t-distributed stochastic neighbor embedding algorithm were employed to study differences in immune patterns. Results: In COVID-19, the immune response consistently showed multiple small-size lymphocyte clusters, suggesting that lymphocyte response is rather modest, possibly due to lymphocytopenia. In H1N1, we found larger lymphocyte clusters that were proximal to large clusters of non-lymphocytes, a possible reflection of increased prevalence of macrophages and other immune cells. Conclusion: Our study shows the potential of computational pathology to uncover immune response features that may not be obvious by routine histopathology visual inspection.

Corredor Germán, Toro Paula, Bera Kaustav, Rasmussen Dylan, Viswanathan Vidya Sankar, Buzzy Christina, Fu Pingfu, Barton Lisa M, Stroberg Edana, Duval Eric, Gilmore Hannah, Mukhopadhyay Sanjay, Madabhushi Anant


H1N1, computational pathology, coronavirus disease 2019, image processing, immune response, machine learning