scholarly journals Fusion of Imaging and Non-Imaging Data for Disease Trajectory Prediction for COVID-19 Patients

2021 ◽  
Author(s):  
Amara Tariq ◽  
Siyi Tang ◽  
Hifza Sakhi ◽  
Leo Anthony Celi ◽  
Janice M. Newsome ◽  
...  
Keyword(s):  
External Validation ◽  
Imaging Features ◽  
Graph Structure ◽  
Imaging Data ◽  
Icu Admission ◽  
Convolutional Networks ◽  
Disease Trajectory ◽  
Multiple Data ◽  
Demographic Similarity ◽  
Discharge From Hospital

ABSTRCATPurposeThis study investigates whether graph-based fusion of imaging data with non-imaging EHR data can improve the prediction of disease trajectory for COVID-19 patients, beyond the prediction performance of only imaging or non-imaging EHR data.Materials and MethodsWe present a novel graph-based framework for fine-grained clinical outcome prediction (discharge, ICU admission, or death) that fuses imaging and non-imaging information using a similarity-based graph structure. Node features are represented by image embedding and edges are encoded with clinical or demographic similarity.ResultsOur experiments on data collected from Emory Healthcare network indicate that our fusion modeling scheme performs consistently better than predictive models using only imaging or non-imaging features, with f1-scores of 0.73, 0.77, and 0.66 for discharge from hospital, mortality, and ICU admission, respectively. External validation was performed on data collected from Mayo Clinic. Our scheme highlights known biases in the model prediction such as bias against patients with alcohol abuse history and bias based on insurance status.ConclusionThe study signifies the importance of fusion of multiple data modalities for accurate prediction of clinical trajectory. Proposed graph structure can model relationships between patients based on non-imaging EHR data and graph convolutional networks can fuse this relationship information with imaging data to effectively predict future disease trajectory more effectively than models employing only imaging or non-imaging data. Forecasting clinical events can enable intelligent resource allocation in hospitals. Our graph-based fusion modeling frameworks can be easily extended to other prediction tasks to efficiently combine imaging data with non-imaging clinical data.

scholarly journals Incorporating radiomics into clinical trials: expert consensus on considerations for data-driven compared to biologically driven quantitative biomarkers

2021 ◽  
Author(s):  
Laure Fournier ◽  
Lena Costaridou ◽  
Luc Bidaut ◽  
Nicolas Michoux ◽  
Frederic E. Lecouvet ◽  
...  
Keyword(s):  
Clinical Trials ◽  
Quantitative Imaging ◽  
A Priori ◽  
External Validation ◽  
Data Driven ◽  
Imaging Biomarkers ◽  
Clinical Validation ◽  
Biological Processes ◽  
Imaging Data ◽  
Selection Of

Abstract Existing quantitative imaging biomarkers (QIBs) are associated with known biological tissue characteristics and follow a well-understood path of technical, biological and clinical validation before incorporation into clinical trials. In radiomics, novel data-driven processes extract numerous visually imperceptible statistical features from the imaging data with no a priori assumptions on their correlation with biological processes. The selection of relevant features (radiomic signature) and incorporation into clinical trials therefore requires additional considerations to ensure meaningful imaging endpoints. Also, the number of radiomic features tested means that power calculations would result in sample sizes impossible to achieve within clinical trials. This article examines how the process of standardising and validating data-driven imaging biomarkers differs from those based on biological associations. Radiomic signatures are best developed initially on datasets that represent diversity of acquisition protocols as well as diversity of disease and of normal findings, rather than within clinical trials with standardised and optimised protocols as this would risk the selection of radiomic features being linked to the imaging process rather than the pathology. Normalisation through discretisation and feature harmonisation are essential pre-processing steps. Biological correlation may be performed after the technical and clinical validity of a radiomic signature is established, but is not mandatory. Feature selection may be part of discovery within a radiomics-specific trial or represent exploratory endpoints within an established trial; a previously validated radiomic signature may even be used as a primary/secondary endpoint, particularly if associations are demonstrated with specific biological processes and pathways being targeted within clinical trials. Key Points • Data-driven processes like radiomics risk false discoveries due to high-dimensionality of the dataset compared to sample size, making adequate diversity of the data, cross-validation and external validation essential to mitigate the risks of spurious associations and overfitting. • Use of radiomic signatures within clinical trials requires multistep standardisation of image acquisition, image analysis and data mining processes. • Biological correlation may be established after clinical validation but is not mandatory.

scholarly journals Development of a machine learning model for predicting pediatric mortality in the early stages of intensive care unit admission

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Bongjin Lee ◽  
Kyunghoon Kim ◽  
Hyejin Hwang ◽  
You Sun Kim ◽  
Eun Hee Chung ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Validation Cohort ◽  
External Validation ◽  
Learning Model ◽  
Derivation Cohort ◽  
Icu Admission ◽  
Early Stages ◽  
Machine Learning Model

AbstractThe aim of this study was to develop a predictive model of pediatric mortality in the early stages of intensive care unit (ICU) admission using machine learning. Patients less than 18 years old who were admitted to ICUs at four tertiary referral hospitals were enrolled. Three hospitals were designated as the derivation cohort for machine learning model development and internal validation, and the other hospital was designated as the validation cohort for external validation. We developed a random forest (RF) model that predicts pediatric mortality within 72 h of ICU admission, evaluated its performance, and compared it with the Pediatric Index of Mortality 3 (PIM 3). The area under the receiver operating characteristic curve (AUROC) of RF model was 0.942 (95% confidence interval [CI] = 0.912–0.972) in the derivation cohort and 0.906 (95% CI = 0.900–0.912) in the validation cohort. In contrast, the AUROC of PIM 3 was 0.892 (95% CI = 0.878–0.906) in the derivation cohort and 0.845 (95% CI = 0.817–0.873) in the validation cohort. The RF model in our study showed improved predictive performance in terms of both internal and external validation and was superior even when compared to PIM 3.

scholarly journals Developing and validating COVID-19 adverse outcome risk prediction models from a bi-national European cohort of 5594 patients

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Espen Jimenez-Solem ◽  
Tonny S. Petersen ◽  
Casper Hansen ◽  
Christian Hansen ◽  
Christina Lioma ◽  
...  
Keyword(s):  
Prediction Models ◽  
Severe Disease ◽  
External Validation ◽  
Area Under The Curve ◽  
Predictive Performance ◽  
Mortality Prediction ◽  
Icu Admission ◽  
Risk Of Death ◽  
Online Risk ◽  
The United Kingdom

AbstractPatients with severe COVID-19 have overwhelmed healthcare systems worldwide. We hypothesized that machine learning (ML) models could be used to predict risks at different stages of management and thereby provide insights into drivers and prognostic markers of disease progression and death. From a cohort of approx. 2.6 million citizens in Denmark, SARS-CoV-2 PCR tests were performed on subjects suspected for COVID-19 disease; 3944 cases had at least one positive test and were subjected to further analysis. SARS-CoV-2 positive cases from the United Kingdom Biobank was used for external validation. The ML models predicted the risk of death (Receiver Operation Characteristics—Area Under the Curve, ROC-AUC) of 0.906 at diagnosis, 0.818, at hospital admission and 0.721 at Intensive Care Unit (ICU) admission. Similar metrics were achieved for predicted risks of hospital and ICU admission and use of mechanical ventilation. Common risk factors, included age, body mass index and hypertension, although the top risk features shifted towards markers of shock and organ dysfunction in ICU patients. The external validation indicated fair predictive performance for mortality prediction, but suboptimal performance for predicting ICU admission. ML may be used to identify drivers of progression to more severe disease and for prognostication patients in patients with COVID-19. We provide access to an online risk calculator based on these findings.

scholarly journals Typical and atypical CT chest imaging findings of novel coronavirus 19 (COVID-19) in correlation with clinical data: impact on the need to ICU admission, ventilation and mortality

2020 ◽  
Vol 51 (1) ◽  
Author(s):  
Doaa M. Emara ◽  
Nagy N. Naguib ◽  
M. A. Moustafa ◽  
Salma M. Ali ◽  
Amr Magdi El Abd
Keyword(s):  
Pleural Effusion ◽  
Complete Recovery ◽  
Clinical Findings ◽  
Mediastinal Lymphadenopathy ◽  
Imaging Features ◽  
Imaging Findings ◽  
Icu Admission ◽  
Mortality Outcome ◽  
Novel Coronavirus ◽  
Contrast Ct

Abstract Background The aim of this study was to highlight the typical and atypical chest CT imaging features at first presentation in 120 patients who were proved to be COVID-19 by PCR and to correlate these findings with the need for ICU admission, ventilation, and mortality. We retrospectively included 120 patients 71 males (59.2%) and 49 females (40.8%) with a mean age of 47.2 ± 14.4 years. Patients subjected to clinical assessment, CBC, PCR for COVID-19, and non-contrast CT chest at first presentation. Typical and atypical imaging findings were reported and correlated with the clinical findings of the patients, the need for ICU admission, ventilation, and mortality. Results Clinically, fever was seen in 112 patients followed by dry cough in 108 patients and malaise in 35 patients. The final outcome was complete recovery in 113 cases and death in 7 cases. Typical CT findings included bilateral peripheral ground-glass opacities (GGO) in 74.7%, multilobar affection in 92.5% while atypical findings such as homogeneous consolidation, pleural effusion, mediastinal lymphadenopathy, and single lobar affection were found in 13.4, 5, 6.7, and 7.5% respectively. A statistically significant association between the presence of white lung, pleural effusion, peripheral GGO, and the need for ICU admission as well as mechanical ventilation was noted. The death was significantly higher among elderly patients; however, no significance was found between the imaging features and mortality. Conclusion CT features at first presentation can predict the need for ICU admission and the need for ventilation but cannot predict the mortality outcome of the patients.

scholarly journals Radiomics in radiation oncology—basics, methods, and limitations

2020 ◽  
Vol 196 (10) ◽  
pp. 848-855
Author(s):  
Philipp Lohmann ◽  
Khaled Bousabarah ◽  
Mauritius Hoevels ◽  
Harald Treuer
Keyword(s):  
Mathematical Models ◽  
Quantitative Imaging ◽  
Human Perception ◽  
Image Features ◽  
Radiotherapy Planning ◽  
Large Field ◽  
Imaging Features ◽  
Imaging Data ◽  
Complete Evaluation ◽  
Vast Number

Abstract Over the past years, the quantity and complexity of imaging data available for the clinical management of patients with solid tumors has increased substantially. Without the support of methods from the field of artificial intelligence (AI) and machine learning, a complete evaluation of the available image information is hardly feasible in clinical routine. Especially in radiotherapy planning, manual detection and segmentation of lesions is laborious, time consuming, and shows significant variability among observers. Here, AI already offers techniques to support radiation oncologists, whereby ultimately, the productivity and the quality are increased, potentially leading to an improved patient outcome. Besides detection and segmentation of lesions, AI allows the extraction of a vast number of quantitative imaging features from structural or functional imaging data that are typically not accessible by means of human perception. These features can be used alone or in combination with other clinical parameters to generate mathematical models that allow, for example, prediction of the response to radiotherapy. Within the large field of AI, radiomics is the subdiscipline that deals with the extraction of quantitative image features as well as the generation of predictive or prognostic mathematical models. This review gives an overview of the basics, methods, and limitations of radiomics, with a focus on patients with brain tumors treated by radiation therapy.

scholarly journals Exploiting an early warning Nomogram for predicting the risk of ICU admission in patients with COVID-19: a multi-center study in China

2020 ◽  
Vol 28 (1) ◽  
Author(s):  
Yiwu Zhou ◽  
Yanqi He ◽  
Huan Yang ◽  
He Yu ◽  
Ting Wang ◽  
...  
Keyword(s):  
Community Health Centers ◽  
Smoking Status ◽  
External Validation ◽  
Individual Risk ◽  
Global Public Health ◽  
Discriminative Ability ◽  
Icu Admission ◽  
Novel Coronavirus ◽  
Individualized Prediction ◽  
Multi Center Study

Abstract Background Novel coronavirus disease 2019 (COVID-19) is a global public health emergency. Here, we developed and validated a practical model based on the data from a multi-center cohort in China for early identification and prediction of which patients will be admitted to the intensive care unit (ICU). Methods Data of 1087 patients with laboratory-confirmed COVID-19 were collected from 49 sites between January 2 and February 28, 2020, in Sichuan and Wuhan. Patients were randomly categorized into the training and validation cohorts (7:3). The least absolute shrinkage and selection operator and logistic regression analyzes were used to develop the nomogram. The performance of the nomogram was evaluated for the C-index, calibration, discrimination, and clinical usefulness. Further, the nomogram was externally validated in a different cohort. Results The individualized prediction nomogram included 6 predictors: age, respiratory rate, systolic blood pressure, smoking status, fever, and chronic kidney disease. The model demonstrated a high discriminative ability in the training cohort (C-index = 0.829), which was confirmed in the external validation cohort (C-index = 0.776). In addition, the calibration plots confirmed good concordance for predicting the risk of ICU admission. Decision curve analysis revealed that the prediction nomogram was clinically useful. Conclusion We established an early prediction model incorporating clinical characteristics that could be quickly obtained on hospital admission, even in community health centers. This model can be conveniently used to predict the individual risk for ICU admission of patients with COVID-19 and optimize the use of limited resources.

scholarly journals NIfTI-MRS: A standard format for magnetic resonance spectroscopic data

2021 ◽  
Author(s):  
William T Clarke ◽  
Mark Mikkelsen ◽  
Georg Oeltzschner ◽  
Tiffany Bell ◽  
Amirmohammad Shamaei ◽  
...  
Keyword(s):  
Data Sharing ◽  
Open Data ◽  
Imaging Data ◽  
Standard Format ◽  
Conversion Point ◽  
Imaging Tool ◽  
Data Formats ◽  
Multiple Data ◽  
Single Voxel ◽  
Online Documentation

Purpose: The use of multiple data formats in the MRS community currently hinders data sharing and integration. NIfTI-MRS is proposed as a standard MR spectroscopy data format, which is implemented as an extension to the neuroimaging informatics technology initiative (NIfTI) format. Using this standardised format will facilitate data sharing, ease algorithm development, and encourage the integration of MRS analysis with other imaging modalities. Methods: A file format based on the NIfTI header extension framework was designed to incorporate essential spectroscopic metadata and additional encoding dimensions. A detailed description of the specification is provided. An open-source command-line conversion program is implemented to enable conversion of single-voxel and spectroscopic imaging data to NIfTI-MRS. To provide visualisation of data in NIfTI-MRS, a dedicated plugin is implemented for FSLeyes, the FSL image viewer. Results: Alongside online documentation, ten example datasets are provided in the proposed format. In addition, minimal examples of NIfTI-MRS readers have been implemented. The conversion software, spec2nii, currently converts fourteen formats to NIfTI-MRS, including DICOM and vendor proprietary formats. Conclusion: The proposed format aims to solve the issue of multiple data formats being used in the MRS community. By providing a single conversion point, it aims to simplify the processing and analysis of MRS data, thereby lowering the barrier to use of MRS. Furthermore, it can serve as the basis for open data sharing, collaboration, and interoperability of analysis programs. It also opens possibility of greater standardisation and harmonisation. By aligning with the dominant format in neuroimaging, NIfTI-MRS enables the use of mature tools present in the imaging community, demonstrated in this work by using a dedicated imaging tool, FSLeyes, as a viewer.

scholarly journals Graph Convolutional Networks using Heat Kernel for Semi-supervised Learning

2019 ◽  
Author(s):  
Bingbing Xu ◽  
Huawei Shen ◽  
Qi Cao ◽  
Keting Cen ◽  
Xueqi Cheng
Keyword(s):  
Heat Kernel ◽  
Supervised Learning ◽  
Low Frequency ◽  
Weighted Average ◽  
Relevant Information ◽  
Semisupervised Learning ◽  
Heat Diffusion ◽  
Graph Structure ◽  
Convolutional Networks ◽  
Benchmark Datasets

Graph convolutional networks gain remarkable success in semi-supervised learning on graph-structured data. The key to graph-based semisupervised learning is capturing the smoothness of labels or features over nodes exerted by graph structure. Previous methods, spectral methods and spatial methods, devote to defining graph convolution as a weighted average over neighboring nodes, and then learn graph convolution kernels to leverage the smoothness to improve the performance of graph-based semi-supervised learning. One open challenge is how to determine appropriate neighborhood that reflects relevant information of smoothness manifested in graph structure. In this paper, we propose GraphHeat, leveraging heat kernel to enhance low-frequency filters and enforce smoothness in the signal variation on the graph. GraphHeat leverages the local structure of target node under heat diffusion to determine its neighboring nodes flexibly, without the constraint of order suffered by previous methods. GraphHeat achieves state-of-the-art results in the task of graph-based semi-supervised classification across three benchmark datasets: Cora, Citeseer and Pubmed.

The Value of Data Mining Technology in the Diagnosis of Bone Metastases from Prostate Cancer Based on Nuclear Medicine Image

2020 ◽  
Vol 10 (7) ◽  
pp. 1660-1668
Author(s):  
Lingmei Wu ◽  
Yan Wei ◽  
Qingyun Wang ◽  
Shuanmeng Ji
Keyword(s):  
Prostate Cancer ◽  
Data Mining ◽  
Nuclear Medicine ◽  
Bone Metastasis ◽  
Image Data ◽  
Background Information ◽  
Imaging Features ◽  
Imaging Data ◽  
Nuclear Medicine Imaging ◽  
Mining Technology

With the continuous development of information construction in the medical industry, a large amount of data related to bone metastasis of prostate cancer can be found in the medical database. It includes a large number of inspection indicators, medical images, and background information such as gender, age, height, weight, and previous medical history. The content is very rich and detailed. The nuclear medicine image processing technology and data mining technology are organically combined to study the feature extraction and loading method of nuclear medicine image data, and the classification method of medical image data, thereby assisting doctors in decision-making diagnosis process and improving accuracy. These have important theoretical significance and broad application prospects. Therefore, based on the nuclear medicine imaging data, this study utilized data mining technology to analyse the nuclear medical imaging data of prostate cancer bone metastasis, and finds and summarizes the imaging features and developmental rules of prostate cancer bone metastasis. So, a BP neural network diagnosis matrix for prostate cancer bone metastasis was constructed. This is valuable and meaningful for the diagnosis, treatment and even medical research of bone metastasis of prostate cancer.

Sign in / Sign up

Export Citation Format

Share Document