An ensemble model for predicting therapeutic response to unfractionated heparin therapy (Preprint)

2021 ◽  
Author(s):  
Ahmad Abdel-Hafez ◽  
Ian A. Scott ◽  
Nazanin Falconer ◽  
Stephen Canaris ◽  
Oscar Bonilla ◽  
...  
Keyword(s):  
Unfractionated Heparin ◽  
Therapeutic Response ◽  
Characteristic Curve ◽  
External Validation ◽  
Heparin Therapy ◽  
Classification Model ◽  
Target Range ◽  
Inpatient Setting ◽  
Episodes Of Care ◽  
Multi Classification

BACKGROUND Unfractionated heparin (UFH), is an anticoagulant drug considered a high-risk medication in that an excessive dose can cause bleeding, while an insufficient dose can lead to a recurrent embolic event. Following initiation of intravenous (IV) UFH, the therapeutic response is monitored using a measure of blood clotting time known as the activated partial thromboplastin time (aPTT). Clinicians iteratively adjust the dose of UFH to a target aPTT range, with the usual therapeutic target range between 60 to 100 seconds. OBJECTIVE The aim of this study was to develop and validate a ML algorithm to predict, aPTT within 12 hours after a specified bolus and maintenance dose of UFH. METHODS This was a retrospective cohort study of 3273 episodes of care from January 2017 to August 2020 using data collected from electronic health records (EHR) of five hospitals in Queensland, Australia. Data from four hospitals were used to build and test ensemble models using cross validation, while the data from the fifth hospital was used for external validation. Modelling was performed using H2O Driverless AI® an automated ML tool, and 17 different experiments were conducted in an iterative process to optimise model accuracy. RESULTS In predicting aPTT, the best performing experiment produced an ensemble with 4x LightGBM models with a root mean square error (RMSE) of 31.35. This dataset was re-purposed as a multi-classification task (sub-therapeutic, therapeutic, and supra-therapeutic aPTT result) and achieved a 59.9% accuracy and area under the receiver operating characteristic curve (AUC) of 0.735. External validation yielded similar results: RMSE of 30.52 +/- 1.29 for the prediction model, and accuracy of 56.8% +/- 3.15 and AUC of 0.724 for the multi-classification model. CONCLUSIONS According to our knowledge, this is the first study of ML applied to IV UFH dosing that has been developed and externally validated in a multisite adult general medical inpatient setting. We present the processes of data collection, preparation, and feature engineering for purposes of replication.

scholarly journals A radiomics-based model for predicting prognosis of locally advanced gastric cancer in the preoperative setting

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Jaeseung Shin ◽  
Joon Seok Lim ◽  
Yong-Min Huh ◽  
Jie-Hyun Kim ◽  
Woo Jin Hyung ◽  
...  
Keyword(s):  
Gastric Cancer ◽  
Advanced Gastric Cancer ◽  
Cox Regression ◽  
Characteristic Curve ◽  
External Validation ◽  
Locally Advanced ◽  
Clinical Model ◽  
Locally Advanced Gastric Cancer ◽  
Contrast Enhanced Ct ◽  
Preoperative Ct

AbstractThis study aims to evaluate the performance of a radiomic signature-based model for predicting recurrence-free survival (RFS) of locally advanced gastric cancer (LAGC) using preoperative contrast-enhanced CT. This retrospective study included a training cohort (349 patients) and an external validation cohort (61 patients) who underwent curative resection for LAGC in 2010 without neoadjuvant therapies. Available preoperative clinical factors, including conventional CT staging and endoscopic data, and 438 radiomic features from the preoperative CT were obtained. To predict RFS, a radiomic model was developed using penalized Cox regression with the least absolute shrinkage and selection operator with ten-fold cross-validation. Internal and external validations were performed using a bootstrapping method. With the final 410 patients (58.2 ± 13.0 years-old; 268 female), the radiomic model consisted of seven selected features. In both of the internal and the external validation, the integrated area under the receiver operating characteristic curve values of both the radiomic model (0.714, P < 0.001 [internal validation]; 0.652, P = 0.010 [external validation]) and the merged model (0.719, P < 0.001; 0.651, P = 0.014) were significantly higher than those of the clinical model (0.616; 0.594). The radiomics-based model on preoperative CT images may improve RFS prediction and high-risk stratification in the preoperative setting of LAGC.

Three plasma-based microRNAs as potent diagnostic biomarkers for endometrial cancer

2021 ◽  
pp. 1-12
Author(s):  
Xingchen Fan ◽  
Minmin Cao ◽  
Cheng Liu ◽  
Cheng Zhang ◽  
Chunyu Li ◽  
...  
Keyword(s):  
Endometrial Cancer ◽  
Operating Characteristic ◽  
Characteristic Curve ◽  
External Validation ◽  
Diagnostic Value ◽  
Diagnostic Biomarkers ◽  
Non Invasive ◽  
Polymerase Chain ◽  
Public Datasets ◽  
Plasma Mirnas

BACKGROUND: MicroRNAs (miRNAs), with noticeable stability and unique expression pattern in plasma of patients with various diseases, are powerful non-invasive biomarkers for cancer detection including endometrial cancer (EC). OBJECTIVE: The objective of this study was to identify promising miRNA biomarkers in plasma to assist the clinical screening of EC. METHODS: A total of 93 EC and 79 normal control (NC) plasma samples were analyzed using Quantitative Real-time Polymerase Chain Reaction (qRT-PCR) in this four-stage experiment. The receiver operating characteristic curve (ROC) analysis was conducted to evaluate the diagnostic value. Additionally, the expression features of the identified miRNAs were further explored in tissues and plasma exosomes samples. RESULTS: The expression of miR-142-3p, miR-146a-5p, and miR-151a-5p was significantly overexpressed in the plasma of EC patients compared with NCs. Areas under the ROC curve of the 3-miRNA signature were 0.729, 0.751, and 0.789 for the training, testing, and external validation phases, respectively. The diagnostic performance of the identified signature proved to be stable in the three public datasets and superior to the other miRNA biomarkers in EC diagnosis. Moreover, the expression of miR-151a-5p was significantly elevated in EC plasma exosomes. CONCLUSIONS: A signature consisting of 3 plasma miRNAs was identified and showed potential for the non-invasive diagnosis of EC.

scholarly journals Development and Verification of a Deep Learning Algorithm to Evaluate Small-Bowel Preparation Quality

Diagnostics ◽  
2021 ◽  
Vol 11 (6) ◽  
pp. 1127
Author(s):  
Ji Hyung Nam ◽  
Dong Jun Oh ◽  
Sumin Lee ◽  
Hyun Joo Song ◽  
Yun Jeong Lim
Keyword(s):  
Deep Learning ◽  
Small Bowel ◽  
Scoring System ◽  
Operating Characteristic ◽  
Clinical Evidence ◽  
Learning Algorithm ◽  
Characteristic Curve ◽  
External Validation ◽  
Test Results ◽  
Deep Learning Algorithm

Capsule endoscopy (CE) quality control requires an objective scoring system to evaluate the preparation of the small bowel (SB). We propose a deep learning algorithm to calculate SB cleansing scores and verify the algorithm’s performance. A 5-point scoring system based on clarity of mucosal visualization was used to develop the deep learning algorithm (400,000 frames; 280,000 for training and 120,000 for testing). External validation was performed using additional CE cases (n = 50), and average cleansing scores (1.0 to 5.0) calculated using the algorithm were compared to clinical grades (A to C) assigned by clinicians. Test results obtained using 120,000 frames exhibited 93% accuracy. The separate CE case exhibited substantial agreement between the deep learning algorithm scores and clinicians’ assessments (Cohen’s kappa: 0.672). In the external validation, the cleansing score decreased with worsening clinical grade (scores of 3.9, 3.2, and 2.5 for grades A, B, and C, respectively, p < 0.001). Receiver operating characteristic curve analysis revealed that a cleansing score cut-off of 2.95 indicated clinically adequate preparation. This algorithm provides an objective and automated cleansing score for evaluating SB preparation for CE. The results of this study will serve as clinical evidence supporting the practical use of deep learning algorithms for evaluating SB preparation quality.

scholarly journals Effectiveness of transfer learning for enhancing tumor classification with a convolutional neural network on frozen sections

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Young-Gon Kim ◽  
Sungchul Kim ◽  
Cristina Eunbee Cho ◽  
In Hye Song ◽  
Hee Jin Lee ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Transfer Learning ◽  
Frozen Section ◽  
Medical Center ◽  
External Validation ◽  
Model Performance ◽  
Classification Model ◽  
Training Dataset

AbstractFast and accurate confirmation of metastasis on the frozen tissue section of intraoperative sentinel lymph node biopsy is an essential tool for critical surgical decisions. However, accurate diagnosis by pathologists is difficult within the time limitations. Training a robust and accurate deep learning model is also difficult owing to the limited number of frozen datasets with high quality labels. To overcome these issues, we validated the effectiveness of transfer learning from CAMELYON16 to improve performance of the convolutional neural network (CNN)-based classification model on our frozen dataset (N = 297) from Asan Medical Center (AMC). Among the 297 whole slide images (WSIs), 157 and 40 WSIs were used to train deep learning models with different dataset ratios at 2, 4, 8, 20, 40, and 100%. The remaining, i.e., 100 WSIs, were used to validate model performance in terms of patch- and slide-level classification. An additional 228 WSIs from Seoul National University Bundang Hospital (SNUBH) were used as an external validation. Three initial weights, i.e., scratch-based (random initialization), ImageNet-based, and CAMELYON16-based models were used to validate their effectiveness in external validation. In the patch-level classification results on the AMC dataset, CAMELYON16-based models trained with a small dataset (up to 40%, i.e., 62 WSIs) showed a significantly higher area under the curve (AUC) of 0.929 than those of the scratch- and ImageNet-based models at 0.897 and 0.919, respectively, while CAMELYON16-based and ImageNet-based models trained with 100% of the training dataset showed comparable AUCs at 0.944 and 0.943, respectively. For the external validation, CAMELYON16-based models showed higher AUCs than those of the scratch- and ImageNet-based models. Model performance for slide feasibility of the transfer learning to enhance model performance was validated in the case of frozen section datasets with limited numbers.

scholarly journals Development and External Validation of Deep-Learning-Based Tumor Grading Models in Soft-Tissue Sarcoma Patients Using MR Imaging

Cancers ◽  
2021 ◽  
Vol 13 (12) ◽  
pp. 2866
Author(s):  
Fernando Navarro ◽  
Hendrik Dapper ◽  
Rebecca Asadpour ◽  
Carolin Knebel ◽  
Matthew B. Spraker ◽  
...  
Keyword(s):  
Deep Learning ◽  
Soft Tissue ◽  
Mr Imaging ◽  
Characteristic Curve ◽  
External Validation ◽  
Soft Tissue Sarcomas ◽  
Treatment Decision ◽  
Receiver Operator Characteristic Curve ◽  
Low Grade ◽  
Tumor Grading

Background: In patients with soft-tissue sarcomas, tumor grading constitutes a decisive factor to determine the best treatment decision. Tumor grading is obtained by pathological work-up after focal biopsies. Deep learning (DL)-based imaging analysis may pose an alternative way to characterize STS tissue. In this work, we sought to non-invasively differentiate tumor grading into low-grade (G1) and high-grade (G2/G3) STS using DL techniques based on MR-imaging. Methods: Contrast-enhanced T1-weighted fat-saturated (T1FSGd) MRI sequences and fat-saturated T2-weighted (T2FS) sequences were collected from two independent retrospective cohorts (training: 148 patients, testing: 158 patients). Tumor grading was determined following the French Federation of Cancer Centers Sarcoma Group in pre-therapeutic biopsies. DL models were developed using transfer learning based on the DenseNet 161 architecture. Results: The T1FSGd and T2FS-based DL models achieved area under the receiver operator characteristic curve (AUC) values of 0.75 and 0.76 on the test cohort, respectively. T1FSGd achieved the best F1-score of all models (0.90). The T2FS-based DL model was able to significantly risk-stratify for overall survival. Attention maps revealed relevant features within the tumor volume and in border regions. Conclusions: MRI-based DL models are capable of predicting tumor grading with good reproducibility in external validation.

External Validation and Test-Retest Reliability of Postpartum Bonding Questionnaire in Spanish Mothers

2021 ◽  
Vol 24 ◽  
Author(s):  
Anna Torres-Giménez ◽  
Alba Roca-Lecumberri ◽  
Bàrbara Sureda ◽  
Susana Andrés-Perpiña ◽  
Bruma Palacios-Hernández ◽  
...  
Keyword(s):  
Mental Health ◽  
Maternal Mental Health ◽  
Characteristic Curve ◽  
External Validation ◽  
Area Under The Curve ◽  
Retest Reliability ◽  
Outpatient Unit ◽  
Auc Value ◽  
Postpartum Bonding ◽  
Test Retest Reliability

Abstract The aim of the present study was to validate the Spanish Postpartum Bonding Questionnaire (PBQ) against external criteria of bonding disorder, as well as to establish its test-retest reliability. One hundred fifty-six postpartum women consecutively recruited from a perinatal mental health outpatient unit completed the PBQ at 4–6 weeks postpartum. Four weeks later, all mothers completed again the PBQ and were interviewed using the Birmingham Interview for Maternal Mental Health to establish the presence of a bonding disorder. Receiver operating characteristic curve analysis revealed an area under the curve (AUC) value for the PBQ total score of 0.93, 95% CI [0.88, 0.98], with the optimal cut-off of 13 for detecting bonding disorders (sensitivity: 92%, specificity: 87%). Optimal cut-off scores for each scale were also obtained. The test-retest reliability coefficients were moderate to good. Our data confirm the validity of PBQ for detecting bonding disorders in Spanish population.

scholarly journals Influence of the Computer-Aided Decision Support System Design on Ultrasound-Based Breast Cancer Classification

Cancers ◽  
2022 ◽  
Vol 14 (2) ◽  
pp. 277
Author(s):  
Zuzanna Anna Magnuska ◽  
Benjamin Theek ◽  
Milita Darguzyte ◽  
Moritz Palmowski ◽  
Elmar Stickeler ◽  
...  
Keyword(s):  
Breast Lesion ◽  
Characteristic Curve ◽  
Cancer Diagnostics ◽  
Lesion Detection ◽  
Machine Learning Algorithms ◽  
Classification Model ◽  
P Value ◽  
Future Design ◽  
Sensitivity Specificity ◽  
Derivatives Of

Automation of medical data analysis is an important topic in modern cancer diagnostics, aiming at robust and reproducible workflows. Therefore, we used a dataset of breast US images (252 malignant and 253 benign cases) to realize and compare different strategies for CAD support in lesion detection and classification. Eight different datasets (including pre-processed and spatially augmented images) were prepared, and machine learning algorithms (i.e., Viola–Jones; YOLOv3) were trained for lesion detection. The radiomics signature (RS) was derived from detection boxes and compared with RS derived from manually obtained segments. Finally, the classification model was established and evaluated concerning accuracy, sensitivity, specificity, and area under the Receiver Operating Characteristic curve. After training on a dataset including logarithmic derivatives of US images, we found that YOLOv3 obtains better results in breast lesion detection (IoU: 0.544 ± 0.081; LE: 0.171 ± 0.009) than the Viola–Jones framework (IoU: 0.399 ± 0.054; LE: 0.096 ± 0.016). Interestingly, our findings show that the classification model trained with RS derived from detection boxes and the model based on the RS derived from a gold standard manual segmentation are comparable (p-value = 0.071). Thus, deriving radiomics signatures from the detection box is a promising technique for building a breast lesion classification model, and may reduce the need for the lesion segmentation step in the future design of CAD systems.

Unfractionated Heparin Therapy in Infants and Children

PEDIATRICS ◽  
2009 ◽  
Vol 123 (3) ◽  
pp. e510-e518 ◽  
Author(s):  
F. Newall ◽  
L. Johnston ◽  
V. Ignjatovic ◽  
P. Monagle
Keyword(s):  
Unfractionated Heparin ◽  
Heparin Therapy ◽  
Infants And Children ◽  
In Infants And Children

scholarly journals Prediction of Outcome After Aneurysmal Subarachnoid Hemorrhage

Stroke ◽  
2019 ◽  
Vol 50 (4) ◽  
pp. 837-844 ◽  
Author(s):  
Carlina E. van Donkelaar ◽  
Nicolaas A. Bakker ◽  
Jaqueline Birks ◽  
Nic J.G.M. Veeger ◽  
Jan D.M. Metzemaekers ◽  
...  
Keyword(s):  
Clinical Outcome ◽  
Operating Characteristic ◽  
Aneurysmal Subarachnoid Hemorrhage ◽  
Characteristic Curve ◽  
External Validation ◽  
World Federation ◽  
Final Model ◽  
Fisher Grade ◽  
Temporal Validation ◽  
Grading Scale

Background and Purpose— Early prediction of clinical outcome after aneurysmal subarachnoid hemorrhage (aSAH) is still lacking accuracy. In this observational cohort study, we aimed to develop and validate an accurate bedside prediction model for clinical outcome after aSAH, to aid decision-making at an early stage. Methods— For the development of the prediction model, a prospectively kept single-center cohort of 1215 aSAH patients, admitted between 1998 and 2014, was used. For temporal validation, a prospective cohort of 224 consecutive aSAH patients from the same center, admitted between 2015 and 2017, was used. External validation was performed using the ISAT (International Subarachnoid Aneurysm Trial) database (2143 patients). Primary outcome measure was poor functional outcome 2 months after aSAH, defined as modified Rankin Scale score 4–6. The model was constructed using multivariate regression analyses. Performance of the model was examined in terms of discrimination and calibration. Results— The final model included 4 predictors independently associated with poor outcome after 2 months: age, World Federation of Neurosurgical Societies grade after resuscitation, aneurysm size, and Fisher grade. Temporal validation showed high discrimination (area under the receiver operating characteristic curve, 0.90; 95% CI, 0.85–0.94), external validation showed fair to good discrimination (area under the receiver operating characteristic curve, 0.73; 95% CI, 0.70–0.76). The model showed satisfactory calibration in both validation cohorts. The SAFIRE grading scale was derived from the final model: size of the aneurysm, age, Fisher grade, world federation of neurosurgical societies after resuscitation. Conclusions— The SAFIRE grading scale is an accurate, generalizable, and easily applicable model for early prediction of clinical outcome after aSAH.

Sign in / Sign up

Export Citation Format

Share Document