1
|
Admiraal MM, Ramos LA, Delgado Olabarriaga S, Marquering HA, Horn J, van Rootselaar AF. Quantitative analysis of EEG reactivity for neurological prognostication after cardiac arrest. Clin Neurophysiol 2021; 132:2240-2247. [PMID: 34315065 DOI: 10.1016/j.clinph.2021.07.004] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/15/2019] [Revised: 04/06/2021] [Accepted: 07/03/2021] [Indexed: 11/28/2022]
Abstract
OBJECTIVE To test whether 1) quantitative analysis of EEG reactivity (EEG-R) using machine learning (ML) is superior to visual analysis, and 2) combining quantitative analyses of EEG-R and EEG background pattern increases prognostic value for prediction of poor outcome after cardiac arrest (CA). METHODS Several types of ML models were trained with twelve quantitative features derived from EEG-R and EEG background data of 134 adult CA patients. Poor outcome was a Cerebral Performance Category score of 3-5 within 6 months. RESULTS The Random Forest (RF) trained on EEG-R showed the highest AUC of 83% (95-CI 80-86) of tested ML classifiers, predicting poor outcome with 46% sensitivity (95%-CI 40-51) and 89% specificity (95%-CI 86-92). Visual analysis of EEG-R had 80% sensitivity and 65% specificity. The RF was also the best classifier for EEG background (AUC 85%, 95%-CI 83-88) at 24 h after CA, with 62% sensitivity (95%-CI 57-67) and 84% specificity (95%-CI 79-88). Combining EEG-R and EEG background RF classifiers reduced the number of false positives. CONCLUSIONS Quantitative EEG-R using ML predicts poor outcome with higher specificity, but lower sensitivity compared to visual analysis of EEG-R, and is of some additional value to ML on EEG background data. SIGNIFICANCE Quantitative EEG-R using ML is a promising alternative to visual analysis and of some added value to ML on EEG background data.
Collapse
Affiliation(s)
- M M Admiraal
- Amsterdam UMC, University of Amsterdam, Department of Neurology/Clinical Neurophysiology, Amsterdam Neuroscience, Amsterdam, the Netherlands.
| | - L A Ramos
- Amsterdam UMC, University of Amsterdam, Department Biomedical Engineering & Physics, Amsterdam Neuroscience, Amsterdam, the Netherlands; Amsterdam UMC, University of Amsterdam, Department of Clinical Epidemiology, Biostatistics and Bioinformatics, Amsterdam, the Netherlands
| | - S Delgado Olabarriaga
- Amsterdam UMC, University of Amsterdam, Department of Clinical Epidemiology, Biostatistics and Bioinformatics, Amsterdam, the Netherlands
| | - H A Marquering
- Amsterdam UMC, University of Amsterdam, Department Biomedical Engineering & Physics, Amsterdam Neuroscience, Amsterdam, the Netherlands; Amsterdam UMC, University of Amsterdam, Department of Radiology and Nuclear Medicine, Amsterdam, the Netherlands
| | - J Horn
- Amsterdam UMC, University of Amsterdam, Laboratory for Experimental Intensive Care and Anesthesiology, Amsterdam, the Netherlands; Amsterdam UMC, University of Amsterdam, Department of Intensive Care, Amsterdam Neuroscience, Amsterdam, the Netherlands
| | - A F van Rootselaar
- Amsterdam UMC, University of Amsterdam, Department of Neurology/Clinical Neurophysiology, Amsterdam Neuroscience, Amsterdam, the Netherlands
| |
Collapse
|
2
|
Kappelhof N, Ramos LA, Kappelhof M, van Os HJA, Chalos V, van Kranendonk KR, Kruyt ND, Roos YBWEM, van Zwam WH, van der Schaaf IC, van Walderveen MAA, Wermer MJH, van Oostenbrugge RJ, Lingsma H, Dippel D, Majoie CBLM, Marquering HA. Evolutionary algorithms and decision trees for predicting poor outcome after endovascular treatment for acute ischemic stroke. Comput Biol Med 2021; 133:104414. [PMID: 33962154 DOI: 10.1016/j.compbiomed.2021.104414] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/27/2020] [Revised: 04/09/2021] [Accepted: 04/15/2021] [Indexed: 11/28/2022]
Abstract
Despite the large overall beneficial effects of endovascular treatment in patients with acute ischemic stroke, severe disability or death still occurs in almost one-third of patients. These patients, who might not benefit from treatment, have been previously identified with traditional logistic regression models, which may oversimplify relations between characteristics and outcome, or machine learning techniques, which may be difficult to interpret. We developed and evaluated a novel evolutionary algorithm for fuzzy decision trees to accurately identify patients with poor outcome after endovascular treatment, which was defined as having a modified Rankin Scale score (mRS) higher or equal to 5. The created decision trees have the benefit of being comprehensible, easily interpretable models, making its predictions easy to explain to patients and practitioners. Insights in the reason for the predicted outcome can encourage acceptance and adaptation in practice and help manage expectations after treatment. We compared our proposed method to CART, the benchmark decision tree algorithm, on classification accuracy and interpretability. The fuzzy decision tree significantly outperformed CART: using 5-fold cross-validation with on average 1090 patients in the training set and 273 patients in the test set, the fuzzy decision tree misclassified on average 77 (standard deviation of 7) patients compared to 83 (±7) using CART. The mean number of nodes (decision and leaf nodes) in the fuzzy decision tree was 11 (±2) compared to 26 (±1) for CART decision trees. With an average accuracy of 72% and much fewer nodes than CART, the developed evolutionary algorithm for fuzzy decision trees might be used to gain insights into the predictive value of patient characteristics and can contribute to the development of more accurate medical outcome prediction methods with improved clarity for practitioners and patients.
Collapse
Affiliation(s)
- N Kappelhof
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - L A Ramos
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands; Department of Clinical Epidemiology and Biostatistics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands.
| | - M Kappelhof
- Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - H J A van Os
- Department of Neurology, Leiden University Medical Center, Leiden, the Netherlands
| | - V Chalos
- Department of Neurology, Erasmus MC - University Medical Center, Rotterdam, the Netherlands; Department of Public Health, Erasmus MC - University Medical Center, Rotterdam, the Netherlands; Department of Radiology and Nuclear Medicine, Erasmus MC - University Medical Center, Rotterdam, the Netherlands
| | - K R van Kranendonk
- Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - N D Kruyt
- Department of Neurology, Leiden University Medical Center, Leiden, the Netherlands
| | - Y B W E M Roos
- Department of Neurology, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - W H van Zwam
- Department of Radiology, Cardiovascular Research Institute Maastricht, Maastricht University Medical Center, Maastricht, the Netherlands
| | - I C van der Schaaf
- Department of Radiology, University Medical Centre, Utrecht, the Netherlands
| | - M A A van Walderveen
- Department of Radiology, Leiden University Medical Center, Leiden, the Netherlands
| | - M J H Wermer
- Department of Neurology, Leiden University Medical Center, Leiden, the Netherlands
| | - R J van Oostenbrugge
- Department of Neurology, Maastricht University Medical Center, Maastricht, the Netherlands
| | - Hester Lingsma
- Department of Public Health, Erasmus MC - University Medical Center, Rotterdam, the Netherlands
| | - Diederik Dippel
- Department of Neurology, Erasmus MC - University Medical Center, Rotterdam, the Netherlands
| | - C B L M Majoie
- Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - H A Marquering
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands; Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| |
Collapse
|
3
|
Núñez-López M, Alarcón Ramos L, Velasco-Hernández JX. Migration rate estimation in an epidemic network. Appl Math Model 2021; 89:1949-1964. [PMID: 32952269 PMCID: PMC7486824 DOI: 10.1016/j.apm.2020.08.025] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 04/25/2020] [Revised: 07/31/2020] [Accepted: 08/06/2020] [Indexed: 05/07/2023]
Abstract
Most of the recent epidemic outbreaks in the world have as a trigger, a strong migratory component as has been evident in the recent Covid-19 pandemic. In this work we address the problem of migration of human populations and its effect on pathogen reinfections in the case of Dengue, using a Markov-chain susceptible-infected-susceptible (SIS) metapopulation model over a network. Our model postulates a general contact rate that represents a local measure of several factors: the population size of infected hosts that arrive at a given location as a function of total population size, the current incidence at neighboring locations, and the connectivity of the network where the disease spreads. This parameter can be interpreted as an indicator of outbreak risk at a given location. This parameter is tied to the fraction of individuals that move across boundaries (migration). To illustrate our model capabilities, we estimate from epidemic Dengue data in Mexico the dynamics of migration at a regional scale incorporating climate variability represented by an index based on precipitation data.
Collapse
Affiliation(s)
- M Núñez-López
- Department of Mathematics, ITAM Río Hondo 1, Ciudad de México 01080, México
| | - L Alarcón Ramos
- Departamento de Matemáticas Aplicadas y Sistemas, Universidad Autónoma Metropolitana, Cuajimalpa, Av. Vasco de Quiroga 4871, Cuajimalpa de Morelos, 05300, México
| | - J X Velasco-Hernández
- Instituto de Matemáticas, Universidad Nacional Autónoma de México, Boulevard Juriquilla No. 3001, Juriquilla, 76230, México
| |
Collapse
|
4
|
van der Steen WE, Marquering HA, Ramos LA, van den Berg R, Coert BA, Boers AMM, Vergouwen MDI, Rinkel GJE, Velthuis BK, Roos YBWEM, Majoie CBLM, Vandertop WP, Verbaan D. Prediction of Outcome Using Quantified Blood Volume in Aneurysmal SAH. AJNR Am J Neuroradiol 2020; 41:1015-1021. [PMID: 32409315 DOI: 10.3174/ajnr.a6575] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/28/2019] [Accepted: 03/26/2020] [Indexed: 11/07/2022]
Abstract
BACKGROUND AND PURPOSE In patients with SAH, the amount of blood is strongly associated with clinical outcome. However, it is commonly estimated with a coarse grading scale, potentially limiting its predictive value. Therefore, we aimed to develop and externally validate prediction models for clinical outcome, including quantified blood volumes, as candidate predictors. MATERIALS AND METHODS Clinical and radiologic candidate predictors were included in a logistic regression model. Unfavorable outcome was defined as a modified Rankin Scale score of 4-6. An automatic hemorrhage-quantification algorithm calculated the total blood volume. Blood was manually classified as cisternal, intraventricular, or intraparenchymal. The model was selected with bootstrapped backward selection and validated with the R 2, C-statistic, and calibration plots. If total blood volume remained in the final model, its performance was compared with models including location-specific blood volumes or the modified Fisher scale. RESULTS The total blood volume, neurologic condition, age, aneurysm size, and history of cardiovascular disease remained in the final models after selection. The externally validated predictive accuracy and discriminative power were high (R 2 = 56% ± 1.8%; mean C-statistic = 0.89 ± 0.01). The location-specific volume models showed a similar performance (R 2 = 56% ± 1%, P = .8; mean C-statistic = 0.89 ± 0.00, P = .4). The modified Fisher models were significantly less accurate (R 2 = 45% ± 3%, P < .001; mean C-statistic = 0.85 ± 0.01, P = .03). CONCLUSIONS The total blood volume-based prediction model for clinical outcome in patients with SAH showed a high predictive accuracy, higher than a prediction model including the commonly used modified Fisher scale.
Collapse
Affiliation(s)
- W E van der Steen
- From the Departments of Biomedical Engineering and Physics (W.E.v.d.S., H.A.M., L.A.R., A.M.M.B.)
- Radiology and Nuclear Medicine (W.E.v.d.S., H.A.M., R.v.d.B., C.B.L.M.M.)
- Neurology (W.E.v.d.S., Y.B.W.E.M.R.)
- Neurosurgical Center Amsterdam (W.E.v.d.S., B.A.C., W.P.V., D.V.), Amsterdam University Medical Center, University of Amsterdam, Amsterdam, the Netherlands
| | - H A Marquering
- From the Departments of Biomedical Engineering and Physics (W.E.v.d.S., H.A.M., L.A.R., A.M.M.B.)
- Radiology and Nuclear Medicine (W.E.v.d.S., H.A.M., R.v.d.B., C.B.L.M.M.)
| | - L A Ramos
- From the Departments of Biomedical Engineering and Physics (W.E.v.d.S., H.A.M., L.A.R., A.M.M.B.)
- Clinical Epidemiology, Biostatistics and Bioinformatics (L.A.R.)
| | - R van den Berg
- Radiology and Nuclear Medicine (W.E.v.d.S., H.A.M., R.v.d.B., C.B.L.M.M.)
| | - B A Coert
- Neurosurgical Center Amsterdam (W.E.v.d.S., B.A.C., W.P.V., D.V.), Amsterdam University Medical Center, University of Amsterdam, Amsterdam, the Netherlands
| | - A M M Boers
- From the Departments of Biomedical Engineering and Physics (W.E.v.d.S., H.A.M., L.A.R., A.M.M.B.)
| | - M D I Vergouwen
- Departments of Neurology and Neurosurgery, Brain Center Rudolf Magnus (M.D.I.V., G.J.E.R.)
| | - G J E Rinkel
- Departments of Neurology and Neurosurgery, Brain Center Rudolf Magnus (M.D.I.V., G.J.E.R.)
| | - B K Velthuis
- Radiology (B.K.V.), University Medical Center, Utrecht University, Utrecht, the Netherlands
| | | | - C B L M Majoie
- Radiology and Nuclear Medicine (W.E.v.d.S., H.A.M., R.v.d.B., C.B.L.M.M.)
| | - W P Vandertop
- Neurosurgical Center Amsterdam (W.E.v.d.S., B.A.C., W.P.V., D.V.), Amsterdam University Medical Center, University of Amsterdam, Amsterdam, the Netherlands
| | - D Verbaan
- Neurosurgical Center Amsterdam (W.E.v.d.S., B.A.C., W.P.V., D.V.), Amsterdam University Medical Center, University of Amsterdam, Amsterdam, the Netherlands
| |
Collapse
|
5
|
Hilbert A, Ramos LA, van Os HJA, Olabarriaga SD, Tolhuisen ML, Wermer MJH, Barros RS, van der Schaaf I, Dippel D, Roos YBWEM, van Zwam WH, Yoo AJ, Emmer BJ, Lycklama À Nijeholt GJ, Zwinderman AH, Strijkers GJ, Majoie CBLM, Marquering HA. Data-efficient deep learning of radiological image data for outcome prediction after endovascular treatment of patients with acute ischemic stroke. Comput Biol Med 2019; 115:103516. [PMID: 31707199 DOI: 10.1016/j.compbiomed.2019.103516] [Citation(s) in RCA: 45] [Impact Index Per Article: 9.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/25/2019] [Revised: 10/15/2019] [Accepted: 10/16/2019] [Indexed: 11/15/2022]
Abstract
Treatment selection is becoming increasingly more important in acute ischemic stroke patient care. Clinical variables and radiological image biomarkers (old age, pre-stroke mRS, NIHSS, occlusion location, ASPECTS, among others) have an important role in treatment selection and prognosis. Radiological biomarkers require expert annotation and are subject to inter-observer variability. Recently, Deep Learning has been introduced to reproduce these radiological image biomarkers. Instead of reproducing these biomarkers, in this work, we investigated Deep Learning techniques for building models to directly predict good reperfusion after endovascular treatment (EVT) and good functional outcome using CT angiography images. These models do not require image annotation and are fast to compute. We compare the Deep Learning models to Machine Learning models using traditional radiological image biomarkers. We explored Residual Neural Network (ResNet) architectures, adapted them with Structured Receptive Fields (RFNN) and auto-encoders (AE) for network weight initialization. We further included model visualization techniques to provide insight into the network's decision-making process. We applied the methods on the MR CLEAN Registry dataset with 1301 patients. The Deep Learning models outperformed the models using traditional radiological image biomarkers in three out of four cross-validation folds for functional outcome (average AUC of 0.71) and for all folds for reperfusion (average AUC of 0.65). Model visualization showed that the arteries were relevant features for functional outcome prediction. The best results were obtained for the ResNet models with RFNN. Auto-encoder initialization often improved the results. We concluded that, in our dataset, automated image analysis with Deep Learning methods outperforms radiological image biomarkers for stroke outcome prediction and has the potential to improve treatment selection.
Collapse
Affiliation(s)
- A Hilbert
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - L A Ramos
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands; Department of Clinical Epidemiology and Biostatistics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands.
| | - H J A van Os
- Department of Neurology, Leiden University Medical Center, Leiden, the Netherlands
| | - S D Olabarriaga
- Department of Clinical Epidemiology and Biostatistics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - M L Tolhuisen
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands; Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - M J H Wermer
- Department of Neurology, Leiden University Medical Center, Leiden, the Netherlands
| | - R S Barros
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - I van der Schaaf
- Department of Radiology and Nuclear Medicine, University Medical Center Utrecht, Utrecht, the Netherlands
| | - D Dippel
- Department of Neurology, Erasmus MC - University Medical Center, Rotterdam, the Netherlands
| | - Y B W E M Roos
- Department of Neurology, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - W H van Zwam
- Department of Radiology and Nuclear Medicine, Maastricht University Medical Center, Maastricht, the Netherlands
| | - A J Yoo
- Neurointervention, Texas Stroke Institute, Dallas-Fort Worth, Texas, USA
| | - B J Emmer
- Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | | | - A H Zwinderman
- Department of Clinical Epidemiology and Biostatistics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - G J Strijkers
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - C B L M Majoie
- Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| | - H A Marquering
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands; Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands
| |
Collapse
|
6
|
Baalman SWE, Schroevers FE, Oakley A, Ramos LA, Lopes RR, Van Der Stuijt W, Brouwer TF, Knops RE, De Groot JR. P2438Deep learning model for atrial fibrillation detection based on single beat morphology. Eur Heart J 2019. [DOI: 10.1093/eurheartj/ehz748.0770] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/14/2022] Open
Abstract
Abstract
Background
The electrocardiogram (ECG) is commonly used, but most recent rhythm discrimination algorithms still lack both specificity and sensitivity. Deep learning techniques have shown promising results in the classification of physiological signals like ECGs.
Purpose
To develop and test a deep learning (DL) model to discriminate between atrial fibrillation (AF) and sinus rhythm (SR).
Methods
For the development of the DL model we used 1499 ECGs sampled at 500 Hz of patients diagnosed with AF. All ECGs were labeled by two experienced investigators. Only ECGs labeled as SR or AF were included in the dataset. To simplify the learning process, solely the first ECG channel was used. The ECG waveforms were preprocessed using the Fourier cosine series to correct for baseline wander. Input data was generated by normalizing and scaling all different heartbeats by centralizing the R peak, leading to 15744 single heart beat samples of 80 data points (figure A). Multiple feedforward architectures were tested with different numbers of layers, filters and activation functions. The models were trained by equally splitting the data (50%SR, 50%AF) in a training (65%), validation (25%) and test set (15%). The best performing model was chosen based on the accuracy.
Results
A total of 1469 ECGs (1061 (72%)SR, 408 (28%)AF) were included. The model with the best performance was a feedforward model consisting three dense layers with ReLU activation and four dense layers with Linear activation. Training of the model was performed in 32 epochs. Validation of the model resulted in an accuracy of 96% (figure B), precision of 95% and recall of 96%.
Conclusions
The morphology based deep learning model developed in this study was able to discriminate atrial fibrillation from sinus rhythm with a fairly high accuracy using a limited size dataset and only one lead.
Collapse
Affiliation(s)
- S W E Baalman
- Amsterdam UMC, University of Amsterdam, Clinical and Experimental Cardiology, Amsterdam, Netherlands (The)
| | | | - A Oakley
- University of Amsterdam, Amsterdam, Netherlands (The)
| | - L A Ramos
- Amsterdam UMC, University of Amsterdam, Clinical and Experimental Cardiology, Amsterdam, Netherlands (The)
| | - R R Lopes
- Amsterdam UMC, University of Amsterdam, Clinical and Experimental Cardiology, Amsterdam, Netherlands (The)
| | - W Van Der Stuijt
- Amsterdam UMC, University of Amsterdam, Clinical and Experimental Cardiology, Amsterdam, Netherlands (The)
| | - T F Brouwer
- Amsterdam UMC, University of Amsterdam, Clinical and Experimental Cardiology, Amsterdam, Netherlands (The)
| | - R E Knops
- Amsterdam UMC, University of Amsterdam, Clinical and Experimental Cardiology, Amsterdam, Netherlands (The)
| | - J R De Groot
- Amsterdam UMC, University of Amsterdam, Clinical and Experimental Cardiology, Amsterdam, Netherlands (The)
| |
Collapse
|
7
|
Lopes RR, van Mourik MS, Schaft EV, Ramos LA, Baan J, Vendrik J, de Mol BAJM, Vis MM, Marquering HA. Value of machine learning in predicting TAVI outcomes. Neth Heart J 2019; 27:443-450. [PMID: 31111457 PMCID: PMC6712116 DOI: 10.1007/s12471-019-1285-7] [Citation(s) in RCA: 17] [Impact Index Per Article: 3.4] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/12/2022] Open
Abstract
BACKGROUND Transcatheter aortic valve implantation (TAVI) has become a commonly applied procedure for high-risk aortic valve stenosis patients. However, for some patients, this procedure does not result in the expected benefits. Previous studies indicated that it is difficult to predict the beneficial effects for specific patients. We aim to study the accuracy of various traditional machine learning (ML) algorithms in the prediction of TAVI outcomes. METHODS AND RESULTS Clinical and laboratory data from 1,478 TAVI patients from a single centre were collected. The outcome measures were improvement of dyspnoea and mortality. Three experiments were performed using (1) screening data, (2) laboratory data, and (3) the combination of both. Five well-established ML techniques were implemented, and the models were evaluated based on the area under the curve (AUC). Random forest classifier achieved the highest AUC (0.70) for predicting mortality. Logistic regression had the highest AUC (0.56) in predicting improvement of dyspnoea. CONCLUSIONS In our single-centre TAVI population, the tree-based models were slightly more accurate than others in predicting mortality. However, ML models performed poorly in predicting improvement of dyspnoea.
Collapse
Affiliation(s)
- R R Lopes
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands
| | - M S van Mourik
- Heart Centre, Amsterdam UMC, University of Amsterdam, Amsterdam Cardiovascular Sciences, Amsterdam, The Netherlands
| | - E V Schaft
- Technical Medicine, University of Twente, Enschede, The Netherlands
| | - L A Ramos
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands.,Department of Clinical Epidemiology, Biostatistics and Bioinformatics, Amsterdam UMC, Amsterdam, The Netherlands
| | - J Baan
- Heart Centre, Amsterdam UMC, University of Amsterdam, Amsterdam Cardiovascular Sciences, Amsterdam, The Netherlands
| | - J Vendrik
- Heart Centre, Amsterdam UMC, University of Amsterdam, Amsterdam Cardiovascular Sciences, Amsterdam, The Netherlands
| | - B A J M de Mol
- Heart Centre, Amsterdam UMC, University of Amsterdam, Amsterdam Cardiovascular Sciences, Amsterdam, The Netherlands
| | - M M Vis
- Heart Centre, Amsterdam UMC, University of Amsterdam, Amsterdam Cardiovascular Sciences, Amsterdam, The Netherlands
| | - H A Marquering
- Department of Biomedical Engineering and Physics, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. .,Department of Radiology and Nuclear Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands.
| |
Collapse
|
8
|
Tsuchiya CT, Kim HSJ, Maximo MFM, Ramos LA. Cost-Minimization Analysis of Bevacizumab Verse Cetuximab in First-Line Treatment for Metastatic Colorectal Cancer in Kras Wild-Type Patients in the Supplementary Health Care System in Brazil. Value Health 2014; 17:A641. [PMID: 27202293 DOI: 10.1016/j.jval.2014.08.2311] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/05/2023]
|
9
|
Defonsi Lestard ME, Ramos LA, Tuttolomondo ME, Ulic SE, Ben Altabef A. Bis (trifluoromethyl) sulfone, CF3SO2CF3: synthesis, vibrational and conformational properties. Spectrochim Acta A Mol Biomol Spectrosc 2012; 96:332-339. [PMID: 22706098 DOI: 10.1016/j.saa.2012.05.049] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/24/2012] [Revised: 05/07/2012] [Accepted: 05/17/2012] [Indexed: 06/01/2023]
Abstract
Bis (trifluoromethyl) sulfone, CF(3)SO(2)CF(3), was obtained as a byproduct in the synthesis of CF(3)SO(2)SCF(3). The compound was characterized by infrared and Raman spectroscopy as well quantum chemical calculations. Quantum mechanical calculations indicate the possible existence of two conformers symmetrically equivalent with C(2) symmetry. The preference for the staggered form was studied using the total energy scheme and the natural bond orbital (NBO) partition scheme. Additionally, the total potential energy was deconvoluted using a sixfold decomposition in terms of a Fourier-type expansion, showing that the hyperconjugative effect was dominant in stabilizing the staggered conformer. Infrared and Raman spectra of CF(3)SO(2)CF(3) were obtained. Harmonic vibrational wavenumbers and a scaled force field were calculated, leading to a final root mean-square deviation of 7.8 cm(-1) when comparing experimental and calculated wavenumbers.
Collapse
Affiliation(s)
- M E Defonsi Lestard
- INQUINOA-CONICET, Instituto de Química Física, Facultad de Bioquímica, Química y Farmacia, Universidad Nacional de Tucumán, San Lorenzo 456, T4000CAN Tucumán, Argentina
| | | | | | | | | |
Collapse
|
10
|
Hidalgo A, Jiménez LPA, Ramos LA, Mroginski MA, Jios JL, Ulic SE, Echeverría GA, Piro OE, Castellano E. Spectroscopic, structural, and conformational properties of (Z)-4,4,4-trifluoro-3-(2-hydroxyethylamino)-1-(2-hydroxyphenyl)-2-buten-1-one, C12H12F3NO3: a trifluoromethyl-substituted β-aminoenone. J Phys Chem A 2012; 116:1110-8. [PMID: 22242788 DOI: 10.1021/jp211536q] [Citation(s) in RCA: 9] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022]
Abstract
The (Z)-4,4,4-trifluoro-3-(2-hydroxyethylamino)-1-(2-hydroxyphenyl)-2-buten-1-one (C(12)H(12)F(3)NO(3)) compound was thoroughly studied by IR, Raman, UV-visible, and (13)C and (19)F NMR spectroscopies. The solid-state molecular structure was determined by X-ray diffraction methods. It crystallizes in the P2(1)/c space group with a = 12.1420(4) Å, b = 7.8210(3) Å, c = 13.8970(5) Å, β = 116.162(2)°, and Z = 4 molecules per unit cell. The molecule shows a nearly planar molecular skeleton, favored by intramolecular OH···O and NH···O bonds, which are arranged in the lattice as an OH···O bonded polymer coiled around crystallographic 2-fold screw-axes. The three postulated tautomers were evaluated using quantum chemical calculations. The lowest energy tautomer (I) calculated with density functional theory methods agrees with the observed crystal structure. The structural and conformational properties are discussed considering the effect of the intra- and intermolecular hydrogen bond interactions.
Collapse
Affiliation(s)
- A Hidalgo
- CEQUINOR (CONICET-UNLP), Facultad de Ciencias Exactas, Universidad Nacional de La Plata, La Plata, República Argentina
| | | | | | | | | | | | | | | | | |
Collapse
|
11
|
Argañaraz ER, Hubbard GB, Ramos LA, Ford AL, Nitz N, Leland MM, Vandeberg JL, Teixeira AR. Blood-sucking lice may disseminate Trypanosoma cruzi infection in baboons. Rev Inst Med Trop Sao Paulo 2001; 43:271-6. [PMID: 11696850 DOI: 10.1590/s0036-46652001000500007] [Citation(s) in RCA: 22] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/22/2022] Open
Abstract
Trypanosoma cruzi (Schyzotrypanum, Chagas, 1909), and Chagas disease are endemic in captive-reared baboons at the Southwest Foundation for Biomedical Research, San Antonio, Texas. We obtained PCR amplification products from DNA extracted from sucking lice collected from the hair and skin of T. cruzi-infected baboons, with specific nested sets of primers for the protozoan kinetoplast DNA, and nuclear DNA. These products were hybridized to their complementary internal sequences. Selected sequences were cloned and sequencing established the presence of T. cruzi nuclear DNA, and minicircle kDNA. Competitive PCR with a kDNA set of primers determined the quantity of approximately 23.9 +/- 18.2 T. cruzi per louse. This finding suggests that the louse may be a vector incidentally contributing to the dissemination of T. cruzi infection in the baboon colony.
Collapse
Affiliation(s)
- E R Argañaraz
- Chagas Disease Multidisciplinary Research Laboratory, Faculty of Medicine, University of Brasília, Brazil
| | | | | | | | | | | | | | | |
Collapse
|
12
|
Ramos LA, Sayer JM, Yagi H, Shah JH, Dipple A, Jerina DM. Effect of benzo-ring hydroxyl groups on site-specific mutagenesis by tetrahydrobenzo[a]pyrene adducts at N(6) of deoxyadenosine. Chem Res Toxicol 2001; 14:1082-9. [PMID: 11511182 DOI: 10.1021/tx010076o] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
Abstract
We have previously investigated the mutations induced on replication in Escherichia coli of the M13mp7L2 genome containing each of the eight possible adducts derived from the four optically active 7,8-diol 9,10-epoxide metabolites of benzo[a]pyrene (B[a]P) by alkylation of a specific deoxyadenosine (dAdo) residue at N(6). Observed mutational frequencies depended in part on the relative spatial orientations of the three hydroxyl groups in these adducts. To determine how the presence or absence of these hydroxyl groups affects mutational response, we have synthesized 16-mer oligonucleotides with the same sequence as one of those previously studied with the diol epoxide adducts, but containing B[a]P-dAdo adducts in which two or all three of the adduct hydroxyl groups were replaced by hydrogen. Transfection of the adducted M13 constructs into SOS-induced Escherichia coli consistently gave fewer infective centers than the control construct, with viabilities ranging from 8.4 to 44.9% relative to control. In general, decreasing the number of adduct hydroxyls decreased the total frequency of substitution mutations induced. For all but one of the present adducts, the total mutational frequency was lower than that for any of the previously reported diol epoxide adducts in the same sequence. Remarkably, this (9S,10R)-adduct with cis orientation of the dAdo residue and the 9-OH group gave the highest mutational frequency of all the B[a]P adducts studied in this sequence, including the diol epoxide adducts. With the present adducts, A --> T transversions predominated, with smaller numbers of A --> G transitions and even fewer A --> C transversions.
Collapse
Affiliation(s)
- L A Ramos
- Chemistry of Carcinogenesis Laboratory, National Cancer Institute at Frederick, Frederick, MD 21702, USA
| | | | | | | | | | | |
Collapse
|
13
|
Teixeira MF, Ramos LA, Fatibello-Filho O, Cavalheiro ET. PbO2-based graphite-epoxy electrode for potentiometric determination of acids and bases in aqueous and aqueous-ethanolic media. Fresenius J Anal Chem 2001; 370:383-6. [PMID: 11495060 DOI: 10.1007/s002160100728] [Citation(s) in RCA: 6] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 10/27/2022]
Abstract
The construction and analytical evaluation of a PbO2-based graphite-epoxy electrode sensitive to H3O+, based on incorporation of lead(IV) oxide in a graphite-epoxy matrix, are described. The data obtained from a variety of acid-base titrations in aqueous and aqueous-ethanolic media were compared with those obtained by use of a glass electrode under the same conditions. The proposed electrode provides a linear response in the pH range from 1 to 11 with a slope of -58.7+/-0.3 mV pH(-1) and -60.8+/-0.2 mV pH(-1) in aqueous and ethanolic media, respectively. The response time was less than 15 s and the lifetime of the electrode was at least eight months (ca. 5000 determinations) and its performance is good in pH determination and end-point detection in potentiometric acid-base titrations in both aqueous and aqueous-ethanolic media.
Collapse
Affiliation(s)
- M F Teixeira
- Departamento de Química, Universidade Federal de São Carlos, SP, Brazil
| | | | | | | |
Collapse
|
14
|
Teixeira MF, Ramos LA, Almeida Neves E, Fatibello-Filho O. Potentiometric determination of acids and bases using a silica gel based carbon-epoxy indicator electrode. Fresenius J Anal Chem 2000; 367:86-9. [PMID: 11227441 DOI: 10.1007/s002160051604] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
Abstract
The construction and the application of a silica gel based carbon-epoxy indicator electrode for the potentiometric determination of acids and bases are described. The effect of composition of silica gel and carbon-epoxy, slope (mV/pH), linear response (pH range) and the use for acid-base titrations were investigated. The data obtained for the acid-base titrations were compared with those obtained using a glass electrode in the same conditions. The electrode showed a linear response in the pH 2 to 13 range with a slope of -40.5 +/- 0.4 mV/pH (at 25 degrees C) and a response time of less than 15 s. The lifetime of the electrode was higher than one year (over 6000 determinations) with a decrease of only 5% of the initial potentiometric response. The silica gel based carbon-epoxy electrode showed excellent results in the end-point indication potentiometric titrations in determination of acids and bases. The miniaturization of the proposed electrode for flow injection analysis was investigated.
Collapse
Affiliation(s)
- M F Teixeira
- Departamento de Química, Universidade Federal de São Carlos, SP, Brazil.
| | | | | | | |
Collapse
|
15
|
Abstract
To determine the mutagenic and genotoxic properties of the major guanine N2-adduct formed by the antitumor drug mitomycin C, we have synthesized a decanucleotide, d(TTACG[MC]TATCT), containing the adduct, which was inserted into a gapped bacteriophage M13 genome. Analysis of the constructed genome indicated that 41% ligation of the adducted 10-mer occurred on both sides of the gap, whereas the control 10-mer ligated with 34% efficiency. After transfection of the adducted single-stranded M13 DNA into Escherichia coli, the adduct was found to be highly genotoxic. Viability of the adducted genome in a repair-competent strain was only 7%, which increased to 12% and 15% upon induction of SOS by irradiating the cells with 254-nm light at 20 and 50 J/m2, respectively. Even lower viability of 2%, 4.6%, and 0.2% was observed in uvrA, uvrB, and uvrC strains, respectively, which increased up to 10-fold with SOS. An examination of the surviving phage populations revealed that the adduct was not detectably mutagenic. No mutants from the repair-proficient strain were detected after analysis of more than 2500 progeny phage. Only 0.2% of the survivors were mutants in the uvrA strain. It is uncertain, however, if they were induced by the adduct, since all the mutants showed untargeted mutations. We conclude that the major guanine N2-adduct formed by mitomycin C is cytotoxic but not appreciably mutagenic in E. coli.
Collapse
Affiliation(s)
- L A Ramos
- Department of Chemistry, University of Connecticut, Storrs 06269, USA
| | | | | | | |
Collapse
|
16
|
Basu AK, Wood ML, Niedernhofer LJ, Ramos LA, Essigmann JM. Mutagenic and genotoxic effects of three vinyl chloride-induced DNA lesions: 1,N6-ethenoadenine, 3,N4-ethenocytosine, and 4-amino-5-(imidazol-2-yl)imidazole. Biochemistry 1993; 32:12793-801. [PMID: 8251500 DOI: 10.1021/bi00210a031] [Citation(s) in RCA: 137] [Impact Index Per Article: 4.4] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/29/2023]
Abstract
The mutagenic and genotoxic properties of 1,N6-ethenoadenine (epsilon Ade), 3,N4-ethenocytosine (epsilon Cyt), and 4-amino-5-(imidazol-2-yl)imidazole (beta) were investigated in vivo. The former two modified bases are known DNA adducts formed by the human carcinogen vinyl chloride; beta is formed by pyrimidine ring-opening of epsilon Ade. Chemically synthesized deoxyhexanucleotides containing epsilon Ade and beta, d[GCT-(epsilon A)GC], and d[GCT(beta)GC], respectively, were described previously [Biochemistry (1987) 26, 5626-5635]. epsilon Cyt was inserted into an oligonucleotide, d[GCTAG(epsilon C)], by a mild enzymatic synthetic procedure, which avoided exposure of the base to alkaline conditions. 3,N4-Etheno-2'-deoxycytidine 3',5'-bisphosphate coupled with reasonable efficiency (30-40%) to the 3'-nucleoside of an acceptor pentamer, d(GCTAG), in a reaction catalyzed by T4 RNA ligase in the presence of ATP. Each of the three modified hexanucleotides and an unmodified control were inserted into a six-base gap positioned at a known site in the genome of bacteriophage M13-NheI. A nick was placed in the DNA strand opposite that containing the single DNA lesions, enabling the formation of singly adducted single-stranded genomes by denaturation. After transfection of the adducted phage DNAs into Escherichia coli, each of the adducts was found to be genotoxic. The most toxic lesion was beta, which reduced survival of the genome by 97%. epsilon Cyt and epsilon Ade reduced survival by 90% and 65%, respectively. An examination of the surviving phage populations revealed that each of the three adducts was mutagenic. The least mutagenic lesion was epsilon Ade (0.1% of the survivors were mutant), which showed primarily A-->G transitions.(ABSTRACT TRUNCATED AT 250 WORDS)
Collapse
Affiliation(s)
- A K Basu
- Department of Chemistry, Massachusetts Institute of Technology, Cambridge 02139
| | | | | | | | | |
Collapse
|
17
|
Ramos LA, Rifkinson N. Acute abdominal manifestations in patients with ventriculo-peritoneal shunts. Bol Asoc Med P R 1990; 82:541-3. [PMID: 2078258] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Subscribe] [Scholar Register] [Indexed: 12/30/2022]
Abstract
Five patients with acute abdominal manifestations after revision of ventriculo-peritoneal shunt were identified. Abdominal pain, nausea, vomiting and distention prompted surgical intervention. Clinical evidence of increased intracranial pressure or shunt malfunction were not prominent findings. Exteriorization of the distal (peritoneal) catherer resolved the acute abdominal findings promptly.
Collapse
Affiliation(s)
- L A Ramos
- Department of Neurological Surgery University of Puerto Rico
| | | |
Collapse
|