Nazanin Esmaili - Academia.edu (original) (raw)

Papers by Nazanin Esmaili

Research paper thumbnail of Machine learning applications to neuroimaging for glioma detection and classification: An artificial intelligence augmented systematic review

Journal of Clinical Neuroscience, Jul 1, 2021

Glioma is the most common primary intraparenchymal tumor of the brain and the 5-year survival rat... more Glioma is the most common primary intraparenchymal tumor of the brain and the 5-year survival rate of high-grade glioma is poor. Magnetic resonance imaging (MRI) is essential for detecting, characterizing and monitoring brain tumors but definitive diagnosis still relies on surgical pathology. Machine learning has been applied to the analysis of MRI data in glioma research and has the potential to change clinical practice and improve patient outcomes. This systematic review synthesizes and analyzes the current state of machine learning applications to glioma MRI data and explores the use of machine learning for systematic review automation. Various datapoints were extracted from the 153 studies that met inclusion criteria and analyzed. Natural language processing (NLP) analysis involved keyword extraction, topic modeling and document classification. Machine learning has been applied to tumor grading and diagnosis, tumor segmentation, non-invasive genomic biomarker identification, detection of progression and patient survival prediction. Model performance was generally strong (AUC=0.87±0.09; sensitivity=0.87±0.10; specificity=0.0.86±0.10; precision=0.88±0.11). Convolutional neural network, support vector machine and random forest algorithms were top performers. Deep learning document classifiers yielded acceptable performance (mean 5-fold cross-validation AUC=0.71). Machine learning tools and data resources were synthesized and summarized to facilitate future research. Machine learning has been widely applied to the processing of MRI data in glioma research and has demonstrated substantial utility. NLP and transfer learning resources enabled the successful development of a replicable method for automating the systematic review article screening process, which has potential for shortening the time from discovery to clinical application in medicine.

Research paper thumbnail of Exact analysis of (<i>R</i>,<i>s</i>,<i>S</i>) inventory control systems with lost sales and zero lead time

Naval Research Logistics, Mar 1, 2019

We study an (R, s, S) inventory control policy with stochastic demand, lost sales, zero lead‐time... more We study an (R, s, S) inventory control policy with stochastic demand, lost sales, zero lead‐time and a target service level to be satisfied. The system is modeled as a discrete time Markov chain for which we present a novel approach to derive exact closed‐form solutions for the limiting distribution of the on‐hand inventory level at the end of a review period, given the reorder level (s) and order‐up‐to level (S). We then establish a relationship between the limiting distributions for adjacent values of the reorder point that is used in an efficient recursive algorithm to determine the optimal parameter values of the (R, s, S) replenishment policy. The algorithm is easy to implement and entails less effort than solving the steady‐state equations for the corresponding Markov model. Point‐of‐use hospital inventory systems share the essential characteristics of the inventory system we model, and a case study using real data from such a system shows that with our approach, optimal policies with significant savings in inventory management effort are easily obtained for a large family of items.

Research paper thumbnail of Predicting Acoustic Hearing Preservation Following Cochlear Implant Surgery Using Machine Learning

Research paper thumbnail of Machine Learning Augmented Interpretation of Chest X-rays: A Systematic Review

Diagnostics

Limitations of the chest X-ray (CXR) have resulted in attempts to create machine learning systems... more Limitations of the chest X-ray (CXR) have resulted in attempts to create machine learning systems to assist clinicians and improve interpretation accuracy. An understanding of the capabilities and limitations of modern machine learning systems is necessary for clinicians as these tools begin to permeate practice. This systematic review aimed to provide an overview of machine learning applications designed to facilitate CXR interpretation. A systematic search strategy was executed to identify research into machine learning algorithms capable of detecting >2 radiographic findings on CXRs published between January 2020 and September 2022. Model details and study characteristics, including risk of bias and quality, were summarized. Initially, 2248 articles were retrieved, with 46 included in the final review. Published models demonstrated strong standalone performance and were typically as accurate, or more accurate, than radiologists or non-radiologist clinicians. Multiple studies d...

Research paper thumbnail of Analysis of Line and Tube Detection Performance of a Chest X-ray Deep Learning Model to Evaluate Hidden Stratification

Diagnostics

This retrospective case-control study evaluated the diagnostic performance of a commercially avai... more This retrospective case-control study evaluated the diagnostic performance of a commercially available chest radiography deep convolutional neural network (DCNN) in identifying the presence and position of central venous catheters, enteric tubes, and endotracheal tubes, in addition to a subgroup analysis of different types of lines/tubes. A held-out test dataset of 2568 studies was sourced from community radiology clinics and hospitals in Australia and the USA, and was then ground-truth labelled for the presence, position, and type of line or tube from the consensus of a thoracic specialist radiologist and an intensive care clinician. DCNN model performance for identifying and assessing the positioning of central venous catheters, enteric tubes, and endotracheal tubes over the entire dataset, as well as within each subgroup, was evaluated. The area under the receiver operating characteristic curve (AUC) was assessed. The DCNN algorithm displayed high performance in detecting the pre...

Research paper thumbnail of The Contextualized Regressive Topic Model

Research paper thumbnail of Predictors of improvement in quality of life at 12-month follow-up in patients undergoing anterior endoscopic skull base surgery

PLOS ONE

Background Patients with pituitary lesions experience decrements in quality of life (QoL) and tre... more Background Patients with pituitary lesions experience decrements in quality of life (QoL) and treatment aims to arrest or improve QoL decline. Objective To detect associations with QoL in trans-nasal endoscopic skull base surgery patients and train supervised learning classifiers to predict QoL improvement at 12 months. Methods A supervised learning analysis of a prospective multi-institutional dataset (451 patients) was conducted. QoL was measured using the anterior skull base surgery questionnaire (ASBS). Factors associated with QoL at baseline and at 12-month follow-up were identified using multivariate logistic regression. Multiple supervised learning models were trained to predict postoperative QoL improvement with five-fold cross-validation. Results ASBS at 12-month follow-up was significantly higher (132.19,SD = 24.87) than preoperative ASBS (121.87,SD = 25.72,p<0.05). High preoperative scores were significantly associated with institution, diabetes and lesions at the plan...

Research paper thumbnail of Effects of a comprehensive brain computed tomography deep-learning model on radiologist detection accuracy: a multireader, multicase study

Background: Non-contrast computed tomography of the brain (NCCTB) is commonly used in clinical pr... more Background: Non-contrast computed tomography of the brain (NCCTB) is commonly used in clinical practice to detect intracranial pathology but is subject to interpretation errors. Machine learning is capable of augmenting clinical decision making and there is an opportunity to apply deep learning to improve the clinical interpretation of NCCTB scans. This retrospective detection accuracy study assessed the performance changes of radiologists assisted by a deep learning model designed to identify many NCCTB clinical findings and also compared the standalone performance of the model with that of unassisted radiologists. Methods: A deep learning model was trained on 212,484 CT scan images of the brain. Thirty-two radiologists each reviewed 2,848 NCCTB cases in a test dataset with and without the assistance of the deep learning model. The consensus of three subspecialist neuroradiologists with access to reports and clinical history was used as a ground truth baseline for comparison. Perfo...

Research paper thumbnail of Neural Topic Model Training with the REBAR Gradient Estimator

Transactions on Asian and Low-Resource Language Information Processing, 2022

Topic modelling is an important approach of unsupervised machine learning that allows automatical... more Topic modelling is an important approach of unsupervised machine learning that allows automatically extracting the main “topics” from large collections of documents. In addition, topic modelling is able to identify the topic proportions of each individual document, which can be helpful for organizing the collections. Many topic modelling algorithms have been proposed to date, including several that leverage advanced techniques such as variational inference and deep autoencoders. However, to date topic modelling has made limited use of reinforcement learning, a framework that has obtained vast success in many other unsupervised learning tasks. For this reason, in this paper we propose training a neural topic model using a reinforcement learning objective, and minimizing the objective with the recently-proposed REBAR gradient estimator. Experiments performed over two probing datasets have shown that the proposed model has achieved improvements over all the compared models in terms of ...

Research paper thumbnail of Charting the potential of brain computed tomography deep learning systems

Journal of Clinical Neuroscience, 2022

Research paper thumbnail of Do comprehensive deep learning algorithms suffer from hidden stratification? A retrospective study on pneumothorax detection in chest radiography

BMJ Open, 2021

ObjectivesTo evaluate the ability of a commercially available comprehensive chest radiography dee... more ObjectivesTo evaluate the ability of a commercially available comprehensive chest radiography deep convolutional neural network (DCNN) to detect simple and tension pneumothorax, as stratified by the following subgroups: the presence of an intercostal drain; rib, clavicular, scapular or humeral fractures or rib resections; subcutaneous emphysema and erect versus non-erect positioning. The hypothesis was that performance would not differ significantly in each of these subgroups when compared with the overall test dataset.DesignA retrospective case–control study was undertaken.SettingCommunity radiology clinics and hospitals in Australia and the USA.ParticipantsA test dataset of 2557 chest radiography studies was ground-truthed by three subspecialty thoracic radiologists for the presence of simple or tension pneumothorax as well as each subgroup other than positioning. Radiograph positioning was derived from radiographer annotations on the images.Outcome measuresDCNN performance for de...

Research paper thumbnail of Assessment of the effect of a comprehensive chest radiograph deep learning model on radiologist reports and patient outcomes: a real-world observational study

BMJ Open, 2021

ObjectivesArtificial intelligence (AI) algorithms have been developed to detect imaging features ... more ObjectivesArtificial intelligence (AI) algorithms have been developed to detect imaging features on chest X-ray (CXR) with a comprehensive AI model capable of detecting 124 CXR findings being recently developed. The aim of this study was to evaluate the real-world usefulness of the model as a diagnostic assistance device for radiologists.DesignThis prospective real-world multicentre study involved a group of radiologists using the model in their daily reporting workflow to report consecutive CXRs and recording their feedback on level of agreement with the model findings and whether this significantly affected their reporting.SettingThe study took place at radiology clinics and hospitals within a large radiology network in Australia between November and December 2020.ParticipantsEleven consultant diagnostic radiologists of varying levels of experience participated in this study.Primary and secondary outcome measuresProportion of CXR cases where use of the AI model led to significant ...

Research paper thumbnail of Systems Engineering Approaches To Understanding and Improving Autism Screening Processes

Research paper thumbnail of A Greedy Primal-Dual Type Heuristic to Select an Inventory Control Policy

Lecture Notes in Management and Industrial Engineering, 2016

We propose a greedy primal-dual type heuristic to jointly optimize the selection of an inventory ... more We propose a greedy primal-dual type heuristic to jointly optimize the selection of an inventory control policy and the allocation of shelf space in order to minimize the expected counting and replenishment costs, while accounting for space limitations. The problem is motivated by an application in the healthcare sector. It addresses the limitations in designing an inventory control system for hospitals stockrooms and the drawbacks of the common approach of using a single policy such as a two-bin Kanban or a Periodic Automatic Replenishment (PAR) system for all items. In the proposed approach, we not only choose policies to use available storage space more efficiently but also consider changing the policies or their parameters to use the space within a selected storage bin more efficiently. On numerical examples where a mathematical programming formulation can be solved in a reasonable amount of time, our experiments indicate that the proposed algorithm is very efficient.

Research paper thumbnail of Topic-Document Inference With the Gumbel-Softmax Distribution

IEEE Access, 2021

Topic modeling is an important application of natural language processing (NLP) that can automati... more Topic modeling is an important application of natural language processing (NLP) that can automatically identify the set of main topics of a given, typically large, collection of documents. In addition to identifying the main topics in the given collection, topic modeling infers which combination of topics is addressed by each individual document (the so-called topic-document inference), which can be useful for their classification and organization. However, the distributional assumptions for this inference are typically restricted to the Dirichlet family which can limit the performance of the model. For this reason, in this paper we propose modeling the topic-document inference with the Gumbel-Softmax distribution, a distribution recently introduced to expand differentiability in deep networks. To set up a performing system, the proposed approach integrates Gumbel-Softmax topic-document inference in a state-of-the-art topic model based on a deep variational autoencoder. Experimental results over two probing datasets show that the proposed approach has been able to outperform the original deep variational autoencoder and other popular topic models in terms of test-set perplexity and two topic coherence measures.

Research paper thumbnail of Machine learning applications to clinical decision support in neurosurgery: an artificial intelligence augmented systematic review

Neurosurgical Review, 2019

Machine learning (ML) involves algorithms learning patterns in large, complex datasets to predict... more Machine learning (ML) involves algorithms learning patterns in large, complex datasets to predict and classify. Algorithms include neural networks (NN), logistic regression (LR), and support vector machines (SVM). ML may generate substantial improvements in neurosurgery. This systematic review assessed the current state of neurosurgical ML applications and the performance of algorithms applied. Our systematic search strategy yielded 6866 results, 70 of which met inclusion criteria. Performance statistics analyzed included area under the receiver operating characteristics curve (AUC), accuracy, sensitivity, and specificity. Natural language processing (NLP) was used to model topics across the corpus and to identify keywords within surgical subspecialties. ML applications were heterogeneous. The densest cluster of studies focused on preoperative evaluation, planning, and outcome prediction in spine surgery. The main algorithms applied were NN, LR, and SVM. Input and output features varied widely and were listed to facilitate future research. The accuracy (F(2,19) = 6.56, p < 0.01) and specificity (F(2,16) = 5.57, p < 0.01) of NN, LR, and SVM differed significantly. NN algorithms demonstrated significantly higher accuracy than LR. SVM demonstrated significantly higher specificity than LR. We found no significant difference between NN, LR, and SVM AUC and sensitivity. NLP topic modeling reached maximum coherence at seven topics, which were defined by modeling approach, surgery type, and pathology themes. Keywords captured research foci within surgical domains. ML technology accurately predicts outcomes and facilitates clinical decision-making in neurosurgery. NNs frequently outperformed other algorithms on supervised learning tasks. This study identified gaps in the literature and opportunities for future neurosurgical ML research.

Research paper thumbnail of Exact analysis of (R , s , S ) inventory control systems with lost sales and zero lead time

Naval Research Logistics (NRL), 2019

We study an (R, s, S) inventory control policy with stochastic demand, lost sales, zero lead‐time... more We study an (R, s, S) inventory control policy with stochastic demand, lost sales, zero lead‐time and a target service level to be satisfied. The system is modeled as a discrete time Markov chain for which we present a novel approach to derive exact closed‐form solutions for the limiting distribution of the on‐hand inventory level at the end of a review period, given the reorder level (s) and order‐up‐to level (S). We then establish a relationship between the limiting distributions for adjacent values of the reorder point that is used in an efficient recursive algorithm to determine the optimal parameter values of the (R, s, S) replenishment policy. The algorithm is easy to implement and entails less effort than solving the steady‐state equations for the corresponding Markov model. Point‐of‐use hospital inventory systems share the essential characteristics of the inventory system we model, and a case study using real data from such a system shows that with our approach, optimal policies with significant savings in inventory management effort are easily obtained for a large family of items.

Research paper thumbnail of Shelf-space optimization models in decentralized automated dispensing cabinets

Operations Research for Health Care, 2018

We propose a mixed integer programming (MIP) model to help clinicians store medications and medic... more We propose a mixed integer programming (MIP) model to help clinicians store medications and medical supplies optimally in space-constrained, decentralized Automated Dispensing Cabinets (ADCs) located on hospital patient floors. We also propose a second MIP model that addresses human errors associated with the selection of pharmaceuticals from floor storage, and not only selects the best set of medications for storage but also determines their optimal layout within the cabinet. To improve the computational performance of these MIP models, we investigate several valid inequalities and relaxations that allow us to solve large, real-world instances in reasonable times. These models are applicable to very general ADCs and are illustrated using real-world data from ADCs at hospitals. Our results indicate that using these models can significantly reduce the time spent by clinical staff on routine logistical functions, while making efficient use of limited space and decreasing risks associated with errors in the selection of medication.

Research paper thumbnail of A Heuristic Approach for Integrated Storage and Shelf-Space Allocation

Lecture Notes in Management and Industrial Engineering, 2015

We address the joint allocation of storage and shelf-space, using an application motivated by the... more We address the joint allocation of storage and shelf-space, using an application motivated by the management of inventory items at Outpatient Clinics (OCs). OCs are limited health care facilities that provide patients with convenient outpatient care within their own community, as opposed to having them visit a major hospital. Currently, patients who are prescribed a prosthetics device during their visit to an OC must often wait for it to be delivered to their homes from a central storage facility. An alternative is the use of integrated storage cabinets at the OCs to store commonly prescribed inventory items that could be given to a patient immediately after a clinic visit. We present, and illustrate with an actual example, a heuristic algorithm for selecting the items to be stocked, along with their shelf space allocations. The objective is to maximize total value based on the desirability of stocking the item for immediate dispensing. The heuristic model considers cabinet characteristics, item size and quantity, and minimum and maximum inventory requirements in order to arrive at the best mix of items and their configuration within the cabinet.

Research paper thumbnail of A REINFORCEd Variational Autoencoder Topic Model

Communications in Computer and Information Science, 2021

Research paper thumbnail of Machine learning applications to neuroimaging for glioma detection and classification: An artificial intelligence augmented systematic review

Journal of Clinical Neuroscience, Jul 1, 2021

Glioma is the most common primary intraparenchymal tumor of the brain and the 5-year survival rat... more Glioma is the most common primary intraparenchymal tumor of the brain and the 5-year survival rate of high-grade glioma is poor. Magnetic resonance imaging (MRI) is essential for detecting, characterizing and monitoring brain tumors but definitive diagnosis still relies on surgical pathology. Machine learning has been applied to the analysis of MRI data in glioma research and has the potential to change clinical practice and improve patient outcomes. This systematic review synthesizes and analyzes the current state of machine learning applications to glioma MRI data and explores the use of machine learning for systematic review automation. Various datapoints were extracted from the 153 studies that met inclusion criteria and analyzed. Natural language processing (NLP) analysis involved keyword extraction, topic modeling and document classification. Machine learning has been applied to tumor grading and diagnosis, tumor segmentation, non-invasive genomic biomarker identification, detection of progression and patient survival prediction. Model performance was generally strong (AUC=0.87±0.09; sensitivity=0.87±0.10; specificity=0.0.86±0.10; precision=0.88±0.11). Convolutional neural network, support vector machine and random forest algorithms were top performers. Deep learning document classifiers yielded acceptable performance (mean 5-fold cross-validation AUC=0.71). Machine learning tools and data resources were synthesized and summarized to facilitate future research. Machine learning has been widely applied to the processing of MRI data in glioma research and has demonstrated substantial utility. NLP and transfer learning resources enabled the successful development of a replicable method for automating the systematic review article screening process, which has potential for shortening the time from discovery to clinical application in medicine.

Research paper thumbnail of Exact analysis of (<i>R</i>,<i>s</i>,<i>S</i>) inventory control systems with lost sales and zero lead time

Naval Research Logistics, Mar 1, 2019

We study an (R, s, S) inventory control policy with stochastic demand, lost sales, zero lead‐time... more We study an (R, s, S) inventory control policy with stochastic demand, lost sales, zero lead‐time and a target service level to be satisfied. The system is modeled as a discrete time Markov chain for which we present a novel approach to derive exact closed‐form solutions for the limiting distribution of the on‐hand inventory level at the end of a review period, given the reorder level (s) and order‐up‐to level (S). We then establish a relationship between the limiting distributions for adjacent values of the reorder point that is used in an efficient recursive algorithm to determine the optimal parameter values of the (R, s, S) replenishment policy. The algorithm is easy to implement and entails less effort than solving the steady‐state equations for the corresponding Markov model. Point‐of‐use hospital inventory systems share the essential characteristics of the inventory system we model, and a case study using real data from such a system shows that with our approach, optimal policies with significant savings in inventory management effort are easily obtained for a large family of items.

Research paper thumbnail of Predicting Acoustic Hearing Preservation Following Cochlear Implant Surgery Using Machine Learning

Research paper thumbnail of Machine Learning Augmented Interpretation of Chest X-rays: A Systematic Review

Diagnostics

Limitations of the chest X-ray (CXR) have resulted in attempts to create machine learning systems... more Limitations of the chest X-ray (CXR) have resulted in attempts to create machine learning systems to assist clinicians and improve interpretation accuracy. An understanding of the capabilities and limitations of modern machine learning systems is necessary for clinicians as these tools begin to permeate practice. This systematic review aimed to provide an overview of machine learning applications designed to facilitate CXR interpretation. A systematic search strategy was executed to identify research into machine learning algorithms capable of detecting >2 radiographic findings on CXRs published between January 2020 and September 2022. Model details and study characteristics, including risk of bias and quality, were summarized. Initially, 2248 articles were retrieved, with 46 included in the final review. Published models demonstrated strong standalone performance and were typically as accurate, or more accurate, than radiologists or non-radiologist clinicians. Multiple studies d...

Research paper thumbnail of Analysis of Line and Tube Detection Performance of a Chest X-ray Deep Learning Model to Evaluate Hidden Stratification

Diagnostics

This retrospective case-control study evaluated the diagnostic performance of a commercially avai... more This retrospective case-control study evaluated the diagnostic performance of a commercially available chest radiography deep convolutional neural network (DCNN) in identifying the presence and position of central venous catheters, enteric tubes, and endotracheal tubes, in addition to a subgroup analysis of different types of lines/tubes. A held-out test dataset of 2568 studies was sourced from community radiology clinics and hospitals in Australia and the USA, and was then ground-truth labelled for the presence, position, and type of line or tube from the consensus of a thoracic specialist radiologist and an intensive care clinician. DCNN model performance for identifying and assessing the positioning of central venous catheters, enteric tubes, and endotracheal tubes over the entire dataset, as well as within each subgroup, was evaluated. The area under the receiver operating characteristic curve (AUC) was assessed. The DCNN algorithm displayed high performance in detecting the pre...

Research paper thumbnail of The Contextualized Regressive Topic Model

Research paper thumbnail of Predictors of improvement in quality of life at 12-month follow-up in patients undergoing anterior endoscopic skull base surgery

PLOS ONE

Background Patients with pituitary lesions experience decrements in quality of life (QoL) and tre... more Background Patients with pituitary lesions experience decrements in quality of life (QoL) and treatment aims to arrest or improve QoL decline. Objective To detect associations with QoL in trans-nasal endoscopic skull base surgery patients and train supervised learning classifiers to predict QoL improvement at 12 months. Methods A supervised learning analysis of a prospective multi-institutional dataset (451 patients) was conducted. QoL was measured using the anterior skull base surgery questionnaire (ASBS). Factors associated with QoL at baseline and at 12-month follow-up were identified using multivariate logistic regression. Multiple supervised learning models were trained to predict postoperative QoL improvement with five-fold cross-validation. Results ASBS at 12-month follow-up was significantly higher (132.19,SD = 24.87) than preoperative ASBS (121.87,SD = 25.72,p<0.05). High preoperative scores were significantly associated with institution, diabetes and lesions at the plan...

Research paper thumbnail of Effects of a comprehensive brain computed tomography deep-learning model on radiologist detection accuracy: a multireader, multicase study

Background: Non-contrast computed tomography of the brain (NCCTB) is commonly used in clinical pr... more Background: Non-contrast computed tomography of the brain (NCCTB) is commonly used in clinical practice to detect intracranial pathology but is subject to interpretation errors. Machine learning is capable of augmenting clinical decision making and there is an opportunity to apply deep learning to improve the clinical interpretation of NCCTB scans. This retrospective detection accuracy study assessed the performance changes of radiologists assisted by a deep learning model designed to identify many NCCTB clinical findings and also compared the standalone performance of the model with that of unassisted radiologists. Methods: A deep learning model was trained on 212,484 CT scan images of the brain. Thirty-two radiologists each reviewed 2,848 NCCTB cases in a test dataset with and without the assistance of the deep learning model. The consensus of three subspecialist neuroradiologists with access to reports and clinical history was used as a ground truth baseline for comparison. Perfo...

Research paper thumbnail of Neural Topic Model Training with the REBAR Gradient Estimator

Transactions on Asian and Low-Resource Language Information Processing, 2022

Topic modelling is an important approach of unsupervised machine learning that allows automatical... more Topic modelling is an important approach of unsupervised machine learning that allows automatically extracting the main “topics” from large collections of documents. In addition, topic modelling is able to identify the topic proportions of each individual document, which can be helpful for organizing the collections. Many topic modelling algorithms have been proposed to date, including several that leverage advanced techniques such as variational inference and deep autoencoders. However, to date topic modelling has made limited use of reinforcement learning, a framework that has obtained vast success in many other unsupervised learning tasks. For this reason, in this paper we propose training a neural topic model using a reinforcement learning objective, and minimizing the objective with the recently-proposed REBAR gradient estimator. Experiments performed over two probing datasets have shown that the proposed model has achieved improvements over all the compared models in terms of ...

Research paper thumbnail of Charting the potential of brain computed tomography deep learning systems

Journal of Clinical Neuroscience, 2022

Research paper thumbnail of Do comprehensive deep learning algorithms suffer from hidden stratification? A retrospective study on pneumothorax detection in chest radiography

BMJ Open, 2021

ObjectivesTo evaluate the ability of a commercially available comprehensive chest radiography dee... more ObjectivesTo evaluate the ability of a commercially available comprehensive chest radiography deep convolutional neural network (DCNN) to detect simple and tension pneumothorax, as stratified by the following subgroups: the presence of an intercostal drain; rib, clavicular, scapular or humeral fractures or rib resections; subcutaneous emphysema and erect versus non-erect positioning. The hypothesis was that performance would not differ significantly in each of these subgroups when compared with the overall test dataset.DesignA retrospective case–control study was undertaken.SettingCommunity radiology clinics and hospitals in Australia and the USA.ParticipantsA test dataset of 2557 chest radiography studies was ground-truthed by three subspecialty thoracic radiologists for the presence of simple or tension pneumothorax as well as each subgroup other than positioning. Radiograph positioning was derived from radiographer annotations on the images.Outcome measuresDCNN performance for de...

Research paper thumbnail of Assessment of the effect of a comprehensive chest radiograph deep learning model on radiologist reports and patient outcomes: a real-world observational study

BMJ Open, 2021

ObjectivesArtificial intelligence (AI) algorithms have been developed to detect imaging features ... more ObjectivesArtificial intelligence (AI) algorithms have been developed to detect imaging features on chest X-ray (CXR) with a comprehensive AI model capable of detecting 124 CXR findings being recently developed. The aim of this study was to evaluate the real-world usefulness of the model as a diagnostic assistance device for radiologists.DesignThis prospective real-world multicentre study involved a group of radiologists using the model in their daily reporting workflow to report consecutive CXRs and recording their feedback on level of agreement with the model findings and whether this significantly affected their reporting.SettingThe study took place at radiology clinics and hospitals within a large radiology network in Australia between November and December 2020.ParticipantsEleven consultant diagnostic radiologists of varying levels of experience participated in this study.Primary and secondary outcome measuresProportion of CXR cases where use of the AI model led to significant ...

Research paper thumbnail of Systems Engineering Approaches To Understanding and Improving Autism Screening Processes

Research paper thumbnail of A Greedy Primal-Dual Type Heuristic to Select an Inventory Control Policy

Lecture Notes in Management and Industrial Engineering, 2016

We propose a greedy primal-dual type heuristic to jointly optimize the selection of an inventory ... more We propose a greedy primal-dual type heuristic to jointly optimize the selection of an inventory control policy and the allocation of shelf space in order to minimize the expected counting and replenishment costs, while accounting for space limitations. The problem is motivated by an application in the healthcare sector. It addresses the limitations in designing an inventory control system for hospitals stockrooms and the drawbacks of the common approach of using a single policy such as a two-bin Kanban or a Periodic Automatic Replenishment (PAR) system for all items. In the proposed approach, we not only choose policies to use available storage space more efficiently but also consider changing the policies or their parameters to use the space within a selected storage bin more efficiently. On numerical examples where a mathematical programming formulation can be solved in a reasonable amount of time, our experiments indicate that the proposed algorithm is very efficient.

Research paper thumbnail of Topic-Document Inference With the Gumbel-Softmax Distribution

IEEE Access, 2021

Topic modeling is an important application of natural language processing (NLP) that can automati... more Topic modeling is an important application of natural language processing (NLP) that can automatically identify the set of main topics of a given, typically large, collection of documents. In addition to identifying the main topics in the given collection, topic modeling infers which combination of topics is addressed by each individual document (the so-called topic-document inference), which can be useful for their classification and organization. However, the distributional assumptions for this inference are typically restricted to the Dirichlet family which can limit the performance of the model. For this reason, in this paper we propose modeling the topic-document inference with the Gumbel-Softmax distribution, a distribution recently introduced to expand differentiability in deep networks. To set up a performing system, the proposed approach integrates Gumbel-Softmax topic-document inference in a state-of-the-art topic model based on a deep variational autoencoder. Experimental results over two probing datasets show that the proposed approach has been able to outperform the original deep variational autoencoder and other popular topic models in terms of test-set perplexity and two topic coherence measures.

Research paper thumbnail of Machine learning applications to clinical decision support in neurosurgery: an artificial intelligence augmented systematic review

Neurosurgical Review, 2019

Machine learning (ML) involves algorithms learning patterns in large, complex datasets to predict... more Machine learning (ML) involves algorithms learning patterns in large, complex datasets to predict and classify. Algorithms include neural networks (NN), logistic regression (LR), and support vector machines (SVM). ML may generate substantial improvements in neurosurgery. This systematic review assessed the current state of neurosurgical ML applications and the performance of algorithms applied. Our systematic search strategy yielded 6866 results, 70 of which met inclusion criteria. Performance statistics analyzed included area under the receiver operating characteristics curve (AUC), accuracy, sensitivity, and specificity. Natural language processing (NLP) was used to model topics across the corpus and to identify keywords within surgical subspecialties. ML applications were heterogeneous. The densest cluster of studies focused on preoperative evaluation, planning, and outcome prediction in spine surgery. The main algorithms applied were NN, LR, and SVM. Input and output features varied widely and were listed to facilitate future research. The accuracy (F(2,19) = 6.56, p < 0.01) and specificity (F(2,16) = 5.57, p < 0.01) of NN, LR, and SVM differed significantly. NN algorithms demonstrated significantly higher accuracy than LR. SVM demonstrated significantly higher specificity than LR. We found no significant difference between NN, LR, and SVM AUC and sensitivity. NLP topic modeling reached maximum coherence at seven topics, which were defined by modeling approach, surgery type, and pathology themes. Keywords captured research foci within surgical domains. ML technology accurately predicts outcomes and facilitates clinical decision-making in neurosurgery. NNs frequently outperformed other algorithms on supervised learning tasks. This study identified gaps in the literature and opportunities for future neurosurgical ML research.

Research paper thumbnail of Exact analysis of (R , s , S ) inventory control systems with lost sales and zero lead time

Naval Research Logistics (NRL), 2019

We study an (R, s, S) inventory control policy with stochastic demand, lost sales, zero lead‐time... more We study an (R, s, S) inventory control policy with stochastic demand, lost sales, zero lead‐time and a target service level to be satisfied. The system is modeled as a discrete time Markov chain for which we present a novel approach to derive exact closed‐form solutions for the limiting distribution of the on‐hand inventory level at the end of a review period, given the reorder level (s) and order‐up‐to level (S). We then establish a relationship between the limiting distributions for adjacent values of the reorder point that is used in an efficient recursive algorithm to determine the optimal parameter values of the (R, s, S) replenishment policy. The algorithm is easy to implement and entails less effort than solving the steady‐state equations for the corresponding Markov model. Point‐of‐use hospital inventory systems share the essential characteristics of the inventory system we model, and a case study using real data from such a system shows that with our approach, optimal policies with significant savings in inventory management effort are easily obtained for a large family of items.

Research paper thumbnail of Shelf-space optimization models in decentralized automated dispensing cabinets

Operations Research for Health Care, 2018

We propose a mixed integer programming (MIP) model to help clinicians store medications and medic... more We propose a mixed integer programming (MIP) model to help clinicians store medications and medical supplies optimally in space-constrained, decentralized Automated Dispensing Cabinets (ADCs) located on hospital patient floors. We also propose a second MIP model that addresses human errors associated with the selection of pharmaceuticals from floor storage, and not only selects the best set of medications for storage but also determines their optimal layout within the cabinet. To improve the computational performance of these MIP models, we investigate several valid inequalities and relaxations that allow us to solve large, real-world instances in reasonable times. These models are applicable to very general ADCs and are illustrated using real-world data from ADCs at hospitals. Our results indicate that using these models can significantly reduce the time spent by clinical staff on routine logistical functions, while making efficient use of limited space and decreasing risks associated with errors in the selection of medication.

Research paper thumbnail of A Heuristic Approach for Integrated Storage and Shelf-Space Allocation

Lecture Notes in Management and Industrial Engineering, 2015

We address the joint allocation of storage and shelf-space, using an application motivated by the... more We address the joint allocation of storage and shelf-space, using an application motivated by the management of inventory items at Outpatient Clinics (OCs). OCs are limited health care facilities that provide patients with convenient outpatient care within their own community, as opposed to having them visit a major hospital. Currently, patients who are prescribed a prosthetics device during their visit to an OC must often wait for it to be delivered to their homes from a central storage facility. An alternative is the use of integrated storage cabinets at the OCs to store commonly prescribed inventory items that could be given to a patient immediately after a clinic visit. We present, and illustrate with an actual example, a heuristic algorithm for selecting the items to be stocked, along with their shelf space allocations. The objective is to maximize total value based on the desirability of stocking the item for immediate dispensing. The heuristic model considers cabinet characteristics, item size and quantity, and minimum and maximum inventory requirements in order to arrive at the best mix of items and their configuration within the cabinet.

Research paper thumbnail of A REINFORCEd Variational Autoencoder Topic Model

Communications in Computer and Information Science, 2021