Parkinson's severity diagnosis explainable model based on 3D multi-head attention residual network.

The severity evaluation of Parkinson's disease (PD) is of great significance for the treatment of PD. However, existing methods either have limitations based on prior knowledge or are invasive methods. To propose a more generalized severity evaluation model, this paper proposes an explainable 3D multi-head attention residual convolution network. First, we introduce the 3D attention-based convolution layer to extract video features. Second, features will be fed into LSTM and residual backbone networks, which can be used to capture the contextual information of the video. Finally, we design a feature compression module to condense the learned contextual features. We develop some interpretable experiments to better explain this black-box model so that it can be better generalized. Experiments show that our model can achieve state-of-the-art diagnosis performance. The proposed lightweight but effective model is expected to serve as a suitable end-to-end deep learning baseline in future research on PD video-based severity evaluation and has the potential for large-scale application in PD telemedicine. The source code is available at https://github.com/JackAILab/MARNet.

Contrast-enhanced CT-based radiomic analysis for determining the response to anti-programmed death-1 therapy in esophageal squamous cell carcinoma patients: A pilot study.

BACKGROUND: In view of the fact that radiomics features have been reported as predictors of immunotherapy to various cancers, this study aimed to develop a prediction model to determine the response to anti-programmed death-1 (anti-PD-1) therapy in esophageal squamous cell carcinoma (ESCC) patients from contrast-enhanced CT (CECT) radiomics features. METHODS: Radiomic analysis of images was performed retrospectively for image samples before and after anti-PD-1 treatment, and efficacy analysis was performed for the results of two different time node evaluations. A total of 68 image samples were included in this study. Quantitative radiomic features were extracted from the images, and the least absolute shrinkage and selection operator method was applied to select radiomic features. After obtaining selected features, three classification models were used to establish a radiomics model to predict the ESCC status and efficacy of therapy. A cross-validation strategy utilizing three folds was employed to train and test the model. Performance evaluation of the model was done using the area under the curve (AUC) of receiver operating characteristic, sensitivity, specificity, and precision metric. RESULTS: Wavelet and area of gray level change (log-sigma) were the most significant radiomic features for predicting therapy efficacy. Fifteen radiomic features from the whole tumor and peritumoral regions were selected and comprised of the fusion radiomics score. A radiomics classification was developed with AUC of 0.82 and 0.884 in the before and after-therapy cohorts, respectively. CONCLUSIONS: The combined model incorporating radiomic features and clinical CECT predictors helps to predict the response to anti-PD-1therapy in patients with ESCC.

A generalizable and interpretable model for mortality risk stratification of sepsis patients in intensive care unit.

PURPOSE: This study aimed to construct a mortality model for the risk stratification of intensive care unit (ICU) patients with sepsis by applying a machine learning algorithm. METHODS: Adult patients who were diagnosed with sepsis during admission to ICU were extracted from MIMIC-III, MIMIC-IV, eICU, and Zigong databases. MIMIC-III was used for model development and internal validation. The other three databases were used for external validation. Our proposed model was developed based on the Extreme Gradient Boosting (XGBoost) algorithm. The generalizability, discrimination, and validation of our model were evaluated. The Shapley Additive Explanation values were used to interpret our model and analyze the contribution of individual features. RESULTS: A total of 16,741, 15,532, 22,617, and 1,198 sepsis patients were extracted from the MIMIC-III, MIMIC-IV, eICU, and Zigong databases, respectively. The proposed model had an area under the receiver operating characteristic curve (AUROC) of 0.84 in the internal validation, which outperformed all the traditional scoring systems. In the external validations, the AUROC was 0.87 in the MIMIC-IV database, better than all the traditional scoring systems; the AUROC was 0.83 in the eICU database, higher than the Simplified Acute Physiology Score III and Sequential Organ Failure Assessment (SOFA),equal to 0.83 of the Acute Physiology and Chronic Health Evaluation IV (APACHE-IV), and the AUROC was 0.68 in the Zigong database, higher than those from the systemic inflammatory response syndrome and SOFA. Furthermore, the proposed model showed the best discriminatory and calibrated capabilities and had the best net benefit in each validation. CONCLUSIONS: The proposed algorithm based on XGBoost and SHAP-value feature selection had high performance in predicting the mortality of sepsis patients within 24 h of ICU admission.

Identifying acute kidney injury subphenotypes using an outcome-driven deep-learning approach.

OBJECTIVE: Acute kidney injury (AKI), a common condition on the intensive-care unit (ICU), is characterized by an abrupt decrease in kidney function within a few hours or days, leading to kidney failure or damage. Although AKI is associated with poor outcomes, current guidelines overlook the heterogeneity among patients with this condition. Identification of AKI subphenotypes could enable targeted interventions and a deeper understanding of the injury's pathophysiology. While previous approaches based on unsupervised representation learning have been used to identify AKI subphenotypes, these methods cannot assess time series or disease severity. METHODS: In this study, we developed a data- and outcome-driven deep-learning (DL) approach to identify and analyze AKI subphenotypes with prognostic and therapeutic implications. Specifically, we developed a supervised long short-term memory (LSTM) autoencoder (AE) with the aim of extracting representation from time-series EHR data that were intricately correlated with mortality. Then, subphenotypes were identified via application of K-means. RESULTS: In two publicly available datasets, three distinct clusters were identified, characterized by mortality rates of 11.3%, 17.3%, and 96.2% in one dataset and 4.6%, 12.1%, and 54.6% in the other. Further analysis demonstrated that AKI subphenotypes identified by our proposed approach were statistically significant on several clinical characteristics and outcomes. CONCLUSION: In this study, our proposed approach could successfully cluster the AKI population in ICU settings into 3 distinct subphenotypes. Thus, such approach could potentially improve outcomes of AKI patients in the ICU, with better risk assessment and potentially better personalized treatment.

Open-source algorithm and software for computed tomography-based virtual pancreatoscopy and other applications.

Pancreatoscopy plays a significant role in the diagnosis and treatment of pancreatic diseases. However, the risk of pancreatoscopy is remarkably greater than that of other endoscopic procedures, such as gastroscopy and bronchoscopy, owing to its severe invasiveness. In comparison, virtual pancreatoscopy (VP) has shown notable advantages. However, because of the low resolution of current computed tomography (CT) technology and the small diameter of the pancreatic duct, VP has limited clinical use. In this study, an optimal path algorithm and super-resolution technique are investigated for the development of an open-source software platform for VP based on 3D Slicer. The proposed segmentation of the pancreatic duct from the abdominal CT images reached an average Dice coefficient of 0.85 with a standard deviation of 0.04. Owing to the excellent segmentation performance, a fly-through visualization of both the inside and outside of the duct was successfully reconstructed, thereby demonstrating the feasibility of VP. In addition, a quantitative analysis of the wall thickness and topology of the duct provides more insight into pancreatic diseases than a fly-through visualization. The entire VP system developed in this study is available at https://github.com/gaoyi/VirtualEndoscopy.git .

Point-of-Care Blood Coagulation Assay Based on Dynamic Monitoring of Blood Viscosity Using Droplet Microfluidics.

Monitoring of the coagulation function has applications in many clinical settings. Routine coagulation assays in the clinic are sample-consuming and slow in turnaround. Microfluidics provides the opportunity to develop coagulation assays that are applicable in point-of-care settings, but reported works required bulky sample pumping units or costly data acquisition instruments. In this work, we developed a microfluidic coagulation assay with a simple setup and easy operation. The device continuously generated droplets of blood sample and buffer mixture and reported the temporal development of blood viscosity during coagulation based on the color appearance of the resultant droplets. We characterized the relationship between blood viscosity and color appearance of the droplets and performed experiments to validate the assay results. In addition, we developed a prototype analyzer equipped with simple fluid pumping and economical imaging module and obtained similar assay measurements. This assay showed great potential to be developed into a point-of-care coagulation test with practical impact.

Acute kidney injury in traumatic brain injury intensive care unit patients.

BACKGROUND: The exact definition of Acute kidney injury (AKI) for patients with traumatic brain injury (TBI) is unknown. AIM: To compare the power of the "Risk, Injury, Failure, Loss of kidney function, and End-stage kidney disease" (RIFLE), Acute Kidney Injury Network (AKIN), Creatinine kinetics (CK), and Kidney Disease Improving Global Outcomes (KDIGO) to determine AKI incidence/stage and their association with the in-hospital mortality rate of patients with TBI. METHODS: This retrospective study collected the data of patients admitted to the intensive care unit for neurotrauma from 2001 to 2012, and 1648 patients were included. The subjects in this study were assessed for the presence and stage of AKI using RIFLE, AKIN, CK, and KDIGO. In addition, the propensity score matching method was used. RESULTS: Among the 1648 patients, 291 (17.7%) had AKI, according to KDIGO. The highest incidence of AKI was found by KDIGO (17.7%), followed by AKIN (17.1%), RIFLE (12.7%), and CK (11.5%) (P = 0.97). Concordance between KDIGO and RIFLE/AKIN/CK was 99.3%/99.1%/99.3% for stage 0, 36.0%/91.5%/44.5% for stage 1, 35.9%/90.6%/11.3% for stage 2, and 47.4%/89.5%/36.8% for stage 3. The in-hospital mortality rates increased with the AKI stage in all four definitions. The severity of AKI by all definitions and stages was not associated with in-hospital mortality in the multivariable analyses (all P > 0.05). CONCLUSION: Differences are seen in AKI diagnosis and in-hospital mortality among the four AKI definitions or stages. This study revealed that KDIGO is the best method to define AKI in patients with TBI.

Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid.

Background: It is a critical challenge to diagnose leptomeningeal metastasis (LM), given its technical difficulty and the lack of typical symptoms. The existing gold standard of diagnosing LM is to use positive cerebrospinal fluid (CSF) cytology, which consumes significantly more time to classify cells under a microscope. Objective: This study aims to establish a deep learning model to classify cancer cells in CSF, thus facilitating doctors to achieve an accurate and fast diagnosis of LM in an early stage. Method: The cerebrospinal fluid laboratory of Xijing Hospital provides 53,255 cells from 90 LM patients in the research. We used two deep convolutional neural networks (CNN) models to classify cells in the CSF. A five-way cell classification model (CNN1) consists of lymphocytes, monocytes, neutrophils, erythrocytes, and cancer cells. A four-way cancer cell classification model (CNN2) consists of lung cancer cells, gastric cancer cells, breast cancer cells, and pancreatic cancer cells. Here, the CNN models were constructed by Resnet-inception-V2. We evaluated the performance of the proposed models on two external datasets and compared them with the results from 42 doctors of various levels of experience in the human-machine tests. Furthermore, we develop a computer-aided diagnosis (CAD) software to generate cytology diagnosis reports in the research rapidly. Results: With respect to the validation set, the mean average precision (mAP) of CNN1 is over 95% and that of CNN2 is close to 80%. Hence, the proposed deep learning model effectively classifies cells in CSF to facilitate the screening of cancer cells. In the human-machine tests, the accuracy of CNN1 is similar to the results from experts, with higher accuracy than doctors in other levels. Moreover, the overall accuracy of CNN2 is 10% higher than that of experts, with a time consumption of only one-third of that consumed by an expert. Using the CAD software saves 90% working time of cytologists. Conclusion: A deep learning method has been developed to assist the LM diagnosis with high accuracy and low time consumption effectively. Thanks to labeled data and step-by-step training, our proposed method can successfully classify cancer cells in the CSF to assist LM diagnosis early. In addition, this unique research can predict cancer's primary source of LM, which relies on cytomorphologic features without immunohistochemistry. Our results show that deep learning can be widely used in medical images to classify cerebrospinal fluid cells. For complex cancer classification tasks, the accuracy of the proposed method is significantly higher than that of specialist doctors, and its performance is better than that of junior doctors and interns. The application of CNNs and CAD software may ultimately aid in expediting the diagnosis and overcoming the shortage of experienced cytologists, thereby facilitating earlier treatment and improving the prognosis of LM.

Validated tool for early prediction of intensive care unit admission in COVID-19 patients.

BACKGROUND: The novel coronavirus disease 2019 (COVID-19) pandemic is a global threat caused by the severe acute respiratory syndrome coronavirus-2. AIM: To develop and validate a risk stratification tool for the early prediction of intensive care unit (ICU) admission among COVID-19 patients at hospital admission. METHODS: The training cohort included COVID-19 patients admitted to the Wuhan Third Hospital. We selected 13 of 65 baseline laboratory results to assess ICU admission risk, which were used to develop a risk prediction model with the random forest (RF) algorithm. A nomogram for the logistic regression model was built based on six selected variables. The predicted models were carefully calibrated, and the predictive performance was evaluated and compared with two previously published models. RESULTS: There were 681 and 296 patients in the training and validation cohorts, respectively. The patients in the training cohort were older than those in the validation cohort (median age: 63.0 vs 49.0 years, P < 0.001), and the percentages of male gender were similar (49.6% vs 49.3%, P = 0.958). The top predictors selected in the RF model were neutrophil-to-lymphocyte ratio, age, lactate dehydrogenase, C-reactive protein, creatinine, D-dimer, albumin, procalcitonin, glucose, platelet, total bilirubin, lactate and creatine kinase. The accuracy, sensitivity and specificity for the RF model were 91%, 88% and 93%, respectively, higher than those for the logistic regression model. The area under the receiver operating characteristic curve of our model was much better than those of two other published methods (0.90 vs 0.82 and 0.75). Model A underestimated risk of ICU admission in patients with a predicted risk less than 30%, whereas the RF risk score demonstrated excellent ability to categorize patients into different risk strata. Our predictive model provided a larger standardized net benefit across the major high-risk range compared with model A. CONCLUSION: Our model can identify ICU admission risk in COVID-19 patients at admission, who can then receive prompt care, thus improving medical resource allocation.

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

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.

Development and validation of a risk stratification model for predicting the mortality of acute kidney injury in critical care patients.

BACKGROUND: This study aimed to develop and validate a model for mortality risk stratification of intensive care unit (ICU) patients with acute kidney injury (AKI) using the machine learning technique. METHODS: Eligible data were extracted from the Medical Information Mart for Intensive Care (MIMIC-III) database. Calibration, discrimination, and risk classification for mortality prediction were evaluated using conventional scoring systems and the new algorithm. A 10-fold cross-validation was performed. The predictive models were externally validated using the eICU database and also patients treated at the Second People's Hospital of Shenzhen between January 2015 to October 2018. RESULTS: For the new model, the areas under the receiver operating characteristic curves (AUROCs) for mortality during hospitalization and at 28 and 90 days after discharge were 0.91, 0.87, and 0.87, respectively, which were higher than for the Simplified Acute Physiology Score (SAPS II) and Sequential Organ Failure Assessment (SOFA). For external validation, the AUROC was 0.82 for in-hospital mortality, higher than SOFA, SAPS II, and Acute Physiology and Chronic Health Evaluation (APACHE) IV in the eICU database, but for the 28- and 90-day mortality, the new model had AUROCs (0.79 and 0.80, respectively) similar to that of SAPS II in the SZ2 database. The reclassification indexes were superior for the new model compared with the conventional scoring systems. CONCLUSIONS: The new risk stratification model shows high performance in predicting mortality in ICU patients with AKI.

Microengineered Flexural Post Rings for Effective Blood Sample Fencing and High-Throughput Measurement of Clot Retraction Force.

During blood clotting, clot retraction alters its mechanical properties and critically affects hemostasis. Despite that, existing clot retraction assays hold limitations such as large footprint and low throughput. In this work, we report the design of flexural post rings for a miniaturized assay of clot retraction force (CRF) with high throughput. Leveraging surface tensions, the post rings hold blood samples in a highly reproducible fashion while simultaneously serving as cantilever beams to measure the CRF. We investigated the effect on the device performance of major parameters, namely, surface hydrophobicity, post number, and post stiffness. We then tested the devices using 14 patient samples and revealed the correlation between CRF and fibrinogen levels. We further implemented an automated liquid handler and developed a high-throughput platform for clot retraction assay. The device's small sample consumption, simple operation, and good compatibility with existing automation facilities make it a promising high-throughput clot retraction assay.