Alireza Akhbardeh
Alireza Akhbardeh has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
- Kouhyar Tavakolian
University of North Dakota, Grand Forks, United States
4 shared publications - Michael A. Jacobs
Department Of Diagnostic And Interventional Imaging McGovern Medical School UTHealth Houston Houston TX USA
4 shared publications - Hersh Sagreiya
Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, USA
3 shared publications - Arash Kamali
Department of Diagnostic and Interventional Imaging, UTHealth Houston, Houston, TX 77030, USA
2 shared publications - Ayse Gamze Yilmaz
Food Science Program, Schmid College of Science and Technology, Chapman University, 1 University Drive, Orange, CA 92866, USA
2 shared publications - Fartash Vasefi
SafetySpect Inc., Grand Forks, ND 58202, USA
2 shared publications
Latest Publications
- Open Access
- 23 June 2026
- Published by Springer Nature in npj Primary Care Respiratory Medicine
- Vol. 36  (1)
Objective dyspnea measurement in real time using a forehead wearable and artificial intelligence
Dyspnea is a complex symptom measured using subjective patient-reported ratings. Continuous, automated dyspnea measurements are needed, especially in critical care and trauma settings with impaired patient communication. We prospectively enrolled 54 pulmonary rehabilitation subjects. Participants completed two treadmill walking trials, during which dyspnea measurements were collected at one-minute intervals automatically using physiologic sensors and patient-reported ratings of perceived breathlessness and exertion (RPB and RPE). The sensor data were used to train machine learning models using either 19 or 7 features to generate an objective dyspnea score (ODS). Classification performance was assessed on a held-out test set, compared against patient-reported RPB and RPE. The model trained on 7 features performed best, resulting in a correlation between predicted and actual scores (percent accuracy) of 0.84 (78.7%) for RPE and 0.86 (83.6%) for RPB. Our system incorporating ODS accurately predicts patients' subjective dyspnea scores and is promising for automated, real-time dyspnea measurement.
- 23 June 2026
- Published by Springer Nature in npj Primary Care Respiratory Medicine
- Vol. 36  (1)
Open AccessObjective dyspnea measurement in real time using a forehead wearable and artificial intelligence
Dyspnea is a complex symptom measured using subjective patient-reported ratings. Continuous, automated dyspnea measurements are needed, especially in critical care and trauma settings with impaired patient communication. We prospectively enrolled 54 pulmonary rehabilitation subjects. Participants completed two treadmill walking trials, during which dyspnea measurements were collected at one-minute intervals automatically using physiologic sensors and patient-reported ratings of perceived breathlessness and exertion (RPB and RPE). The sensor data were used to train machine learning models using either 19 or 7 features to generate an objective dyspnea score (ODS). Classification performance was assessed on a held-out test set, compared against patient-reported RPB and RPE. The model trained on 7 features performed best, resulting in a correlation between predicted and actual scores (percent accuracy) of 0.84 (78.7%) for RPE and 0.86 (83.6%) for RPB. Our system incorporating ODS accurately predicts patients' subjective dyspnea scores and is promising for automated, real-time dyspnea measurement.
- Open Access
An Unsupervised Approach for Treatment Effectiveness Monitoring Using Curvature Learning
Contrast-enhanced ultrasound could assess whether cancer chemotherapeutic agents work in days, rather than waiting 2–3 months, as is typical using the Response Evaluation Criteria in Solid Tumors (RECIST), therefore avoiding toxic side effects and expensive, ineffective therapy. A total of 40 mice were implanted with human colon cancer cells: treatment-sensitive mice in control (n = 10, receiving saline) and treated (n = 10, receiving bevacizumab) groups and treatment-resistant mice in control (n = 10) and treated (n = 10) groups. Each mouse was imaged using 3D dynamic contrast-enhanced ultrasound with Definity microbubbles. Curvature learning, an unsupervised learning approach, quantized pixels into three classes—blue, yellow, and red—representing normal, intermediate, and high cancer probability, both at baseline and after treatment. Next, a curvature learning score was calculated for each mouse using statistical measures representing variations in these three color classes across each frame from cine ultrasound images obtained during contrast administration on a given day (intra-day variability) and between pre- and post-treatment days (inter-day variability). A Wilcoxon rank-sum test compared score distributions between treated, treatment-sensitive mice and all others. There was a statistically significant difference in tumor score between the treated, treatment-sensitive group (n = 10) and all others (n = 30) (p = 0.0051). Curvature learning successfully identified treatment response, detecting changes in tumor perfusion before changes in tumor size. A similar technique could be developed for humans.
Open AccessAn Unsupervised Approach for Treatment Effectiveness Monitoring Using Curvature Learning
Contrast-enhanced ultrasound could assess whether cancer chemotherapeutic agents work in days, rather than waiting 2–3 months, as is typical using the Response Evaluation Criteria in Solid Tumors (RECIST), therefore avoiding toxic side effects and expensive, ineffective therapy. A total of 40 mice were implanted with human colon cancer cells: treatment-sensitive mice in control (n = 10, receiving saline) and treated (n = 10, receiving bevacizumab) groups and treatment-resistant mice in control (n = 10) and treated (n = 10) groups. Each mouse was imaged using 3D dynamic contrast-enhanced ultrasound with Definity microbubbles. Curvature learning, an unsupervised learning approach, quantized pixels into three classes—blue, yellow, and red—representing normal, intermediate, and high cancer probability, both at baseline and after treatment. Next, a curvature learning score was calculated for each mouse using statistical measures representing variations in these three color classes across each frame from cine ultrasound images obtained during contrast administration on a given day (intra-day variability) and between pre- and post-treatment days (inter-day variability). A Wilcoxon rank-sum test compared score distributions between treated, treatment-sensitive mice and all others. There was a statistically significant difference in tumor score between the treated, treatment-sensitive group (n = 10) and all others (n = 30) (p = 0.0051). Curvature learning successfully identified treatment response, detecting changes in tumor perfusion before changes in tumor size. A similar technique could be developed for humans.
- Open Access
The Use of Apparent Diffusion Coefficient Values for Differentiating Bevacizumab-Related Cytotoxicity from Tumor Recurrence and Radiation Necrosis in Glioblastoma
Objectives: Glioblastomas (GBM) are the most common primary invasive neoplasms of the brain. Distinguishing between lesion recurrence and different types of treatment related changes in patients with GBM remains challenging using conventional MRI imaging techniques. Therefore, accurate and precise differentiation between true progression or pseudoresponse is crucial in deciding on the appropriate course of treatment. This retrospective study investigated the potential of apparent diffusion coefficient (ADC) map values derived from diffusion-weighted imaging (DWI) as a noninvasive method to increase diagnostic accuracy in treatment response. Methods: A cohort of 21 glioblastoma patients (mean age: 59.2 ± 11.8, 12 Male, 9 Female) that underwent treatment with bevacizumab were selected. The ADC values were calculated from the DWI images obtained from a standardized brain protocol across 1.5-T and 3-T MRI scanners. Ratios were calculated for rADC values. Lesions were classified as bevacizumab-induced cytotoxicity based on characteristic imaging features (well-defined regions of restricted diffusion with persistent diffusion restriction over the course of weeks without tissue volume loss and absence of contrast enhancement). The rADC value was compared to these values in radiation necrosis and recurrent lesions, which were concluded in our prior study. The nonparametric Wilcoxon signed rank test with p < 0.05 was used for significance. Results: The mean ± SD age of the selected patients was 59.2 ± 11.8. ADC values and corresponding mean rADC values for bevacizumab-induced cytotoxicity were 248.1 ± 67.2 and 0.39 ± 0.10, respectively. These results were compared to the ADC values and corresponding mean rADC values of tumor progression and radiation necrosis. Significant differences between rADC values were observed in all three groups (p < 0.001). Bevacizumab-induced cytotoxicity had statistically significant lower ADC values compared to both tumor recurrence and radiation necrosis. Conclusion: The study demonstrates the potential of ADC values as noninvasive imaging biomarkers for differentiating recurrent glioblastoma from radiation necrosis and bevacizumab-induced cytotoxicity.
Open AccessThe Use of Apparent Diffusion Coefficient Values for Differentiating Bevacizumab-Related Cytotoxicity from Tumor Recurrence and Radiation Necrosis in Glioblastoma
Objectives: Glioblastomas (GBM) are the most common primary invasive neoplasms of the brain. Distinguishing between lesion recurrence and different types of treatment related changes in patients with GBM remains challenging using conventional MRI imaging techniques. Therefore, accurate and precise differentiation between true progression or pseudoresponse is crucial in deciding on the appropriate course of treatment. This retrospective study investigated the potential of apparent diffusion coefficient (ADC) map values derived from diffusion-weighted imaging (DWI) as a noninvasive method to increase diagnostic accuracy in treatment response. Methods: A cohort of 21 glioblastoma patients (mean age: 59.2 ± 11.8, 12 Male, 9 Female) that underwent treatment with bevacizumab were selected. The ADC values were calculated from the DWI images obtained from a standardized brain protocol across 1.5-T and 3-T MRI scanners. Ratios were calculated for rADC values. Lesions were classified as bevacizumab-induced cytotoxicity based on characteristic imaging features (well-defined regions of restricted diffusion with persistent diffusion restriction over the course of weeks without tissue volume loss and absence of contrast enhancement). The rADC value was compared to these values in radiation necrosis and recurrent lesions, which were concluded in our prior study. The nonparametric Wilcoxon signed rank test with p < 0.05 was used for significance. Results: The mean ± SD age of the selected patients was 59.2 ± 11.8. ADC values and corresponding mean rADC values for bevacizumab-induced cytotoxicity were 248.1 ± 67.2 and 0.39 ± 0.10, respectively. These results were compared to the ADC values and corresponding mean rADC values of tumor progression and radiation necrosis. Significant differences between rADC values were observed in all three groups (p < 0.001). Bevacizumab-induced cytotoxicity had statistically significant lower ADC values compared to both tumor recurrence and radiation necrosis. Conclusion: The study demonstrates the potential of ADC values as noninvasive imaging biomarkers for differentiating recurrent glioblastoma from radiation necrosis and bevacizumab-induced cytotoxicity.
- Open Access
- 16 March 2024
- Published by MDPI AG in Diagnostics
- Vol. 14  (6) , 632
Automatic Active Lesion Tracking in Multiple Sclerosis Using Unsupervised Machine Learning
Background: Identifying active lesions in magnetic resonance imaging (MRI) is crucial for the diagnosis and treatment planning of multiple sclerosis (MS). Active lesions on MRI are identified following the administration of Gadolinium-based contrast agents (GBCAs). However, recent studies have reported that repeated administration of GBCA results in the accumulation of Gd in tissues. In addition, GBCA administration increases health care costs. Thus, reducing or eliminating GBCA administration for active lesion detection is important for improved patient safety and reduced healthcare costs. Current state-of-the-art methods for identifying active lesions in brain MRI without GBCA administration utilize data-intensive deep learning methods. Objective: To implement nonlinear dimensionality reduction (NLDR) methods, locally linear embedding (LLE) and isometric feature mapping (Isomap), which are less data-intensive, for automatically identifying active lesions on brain MRI in MS patients, without the administration of contrast agents. Materials and Methods: Fluid-attenuated inversion recovery (FLAIR), T2-weighted, proton density-weighted, and pre- and post-contrast T1-weighted images were included in the multiparametric MRI dataset used in this study. Subtracted pre- and post-contrast T1-weighted images were labeled by experts as active lesions (ground truth). Unsupervised methods, LLE and Isomap, were used to reconstruct multiparametric brain MR images into a single embedded image. Active lesions were identified on the embedded images and compared with ground truth lesions. The performance of NLDR methods was evaluated by calculating the Dice similarity (DS) index between the observed and identified active lesions in embedded images. Results: LLE and Isomap, were applied to 40 MS patients, achieving median DS scores of 0.74 ± 0.1 and 0.78 ± 0.09, respectively, outperforming current state-of-the-art methods. Conclusions: NLDR methods, Isomap and LLE, are viable options for the identification of active MS lesions on non-contrast images, and potentially could be used as a clinical decision tool.
- 16 March 2024
- Published by MDPI AG in Diagnostics
- Vol. 14  (6) , 632
Open AccessAutomatic Active Lesion Tracking in Multiple Sclerosis Using Unsupervised Machine Learning
Background: Identifying active lesions in magnetic resonance imaging (MRI) is crucial for the diagnosis and treatment planning of multiple sclerosis (MS). Active lesions on MRI are identified following the administration of Gadolinium-based contrast agents (GBCAs). However, recent studies have reported that repeated administration of GBCA results in the accumulation of Gd in tissues. In addition, GBCA administration increases health care costs. Thus, reducing or eliminating GBCA administration for active lesion detection is important for improved patient safety and reduced healthcare costs. Current state-of-the-art methods for identifying active lesions in brain MRI without GBCA administration utilize data-intensive deep learning methods. Objective: To implement nonlinear dimensionality reduction (NLDR) methods, locally linear embedding (LLE) and isometric feature mapping (Isomap), which are less data-intensive, for automatically identifying active lesions on brain MRI in MS patients, without the administration of contrast agents. Materials and Methods: Fluid-attenuated inversion recovery (FLAIR), T2-weighted, proton density-weighted, and pre- and post-contrast T1-weighted images were included in the multiparametric MRI dataset used in this study. Subtracted pre- and post-contrast T1-weighted images were labeled by experts as active lesions (ground truth). Unsupervised methods, LLE and Isomap, were used to reconstruct multiparametric brain MR images into a single embedded image. Active lesions were identified on the embedded images and compared with ground truth lesions. The performance of NLDR methods was evaluated by calculating the Dice similarity (DS) index between the observed and identified active lesions in embedded images. Results: LLE and Isomap, were applied to 40 MS patients, achieving median DS scores of 0.74 ± 0.1 and 0.78 ± 0.09, respectively, outperforming current state-of-the-art methods. Conclusions: NLDR methods, Isomap and LLE, are viable options for the identification of active MS lesions on non-contrast images, and potentially could be used as a clinical decision tool.
- Top Cited PapersOpen Access
- 6 October 2023
- Published by MDPI AG in Diagnostics
- Vol. 13  (19) , 3140
Medical Imaging Applications of Federated Learning
Since its introduction in 2016, researchers have applied the idea of Federated Learning (FL) to several domains ranging from edge computing to banking. The technique’s inherent security benefits, privacy-preserving capabilities, ease of scalability, and ability to transcend data biases have motivated researchers to use this tool on healthcare datasets. While several reviews exist detailing FL and its applications, this review focuses solely on the different applications of FL to medical imaging datasets, grouping applications by diseases, modality, and/or part of the body. This Systematic Literature review was conducted by querying and consolidating results from ArXiv, IEEE Xplorer, and PubMed. Furthermore, we provide a detailed description of FL architecture, models, descriptions of the performance achieved by FL models, and how results compare with traditional Machine Learning (ML) models. Additionally, we discuss the security benefits, highlighting two primary forms of privacy-preserving techniques, including homomorphic encryption and differential privacy. Finally, we provide some background information and context regarding where the contributions lie. The background information is organized into the following categories: architecture/setup type, data-related topics, security, and learning types. While progress has been made within the field of FL and medical imaging, much room for improvement and understanding remains, with an emphasis on security and data issues remaining the primary concerns for researchers. Therefore, improvements are constantly pushing the field forward. Finally, we highlighted the challenges in deploying FL in medical imaging applications and provided recommendations for future directions.
- 6 October 2023
- Published by MDPI AG in Diagnostics
- Vol. 13  (19) , 3140
Top Cited PapersOpen AccessMedical Imaging Applications of Federated Learning
Since its introduction in 2016, researchers have applied the idea of Federated Learning (FL) to several domains ranging from edge computing to banking. The technique’s inherent security benefits, privacy-preserving capabilities, ease of scalability, and ability to transcend data biases have motivated researchers to use this tool on healthcare datasets. While several reviews exist detailing FL and its applications, this review focuses solely on the different applications of FL to medical imaging datasets, grouping applications by diseases, modality, and/or part of the body. This Systematic Literature review was conducted by querying and consolidating results from ArXiv, IEEE Xplorer, and PubMed. Furthermore, we provide a detailed description of FL architecture, models, descriptions of the performance achieved by FL models, and how results compare with traditional Machine Learning (ML) models. Additionally, we discuss the security benefits, highlighting two primary forms of privacy-preserving techniques, including homomorphic encryption and differential privacy. Finally, we provide some background information and context regarding where the contributions lie. The background information is organized into the following categories: architecture/setup type, data-related topics, security, and learning types. While progress has been made within the field of FL and medical imaging, much room for improvement and understanding remains, with an emphasis on security and data issues remaining the primary concerns for researchers. Therefore, improvements are constantly pushing the field forward. Finally, we highlighted the challenges in deploying FL in medical imaging applications and provided recommendations for future directions.
- Open Access
- 16 August 2023
- Published by MDPI AG in Diagnostics
- Vol. 13  (16) , 2692
Automated Lung Ultrasound Pulmonary Disease Quantification Using an Unsupervised Machine Learning Technique for COVID-19
COVID-19 is an ongoing global health pandemic. Although COVID-19 can be diagnosed with various tests such as PCR, these tests do not establish pulmonary disease burden. Whereas point-of-care lung ultrasound (POCUS) can directly assess the severity of characteristic pulmonary findings of COVID-19, the advantage of using US is that it is inexpensive, portable, and widely available for use in many clinical settings. For automated assessment of pulmonary findings, we have developed an unsupervised learning technique termed the calculated lung ultrasound (CLU) index. The CLU can quantify various types of lung findings, such as A or B lines, consolidations, and pleural effusions, and it uses these findings to calculate a CLU index score, which is a quantitative measure of pulmonary disease burden. This is accomplished using an unsupervised, patient-specific approach that does not require training on a large dataset. The CLU was tested on 52 lung ultrasound examinations from several institutions. CLU demonstrated excellent concordance with radiologist findings in different pulmonary disease states. Given the global nature of COVID-19, the CLU would be useful for sonographers and physicians in resource-strapped areas with limited ultrasound training and diagnostic capacities for more accurate assessment of pulmonary status.
- 16 August 2023
- Published by MDPI AG in Diagnostics
- Vol. 13  (16) , 2692
Open AccessAutomated Lung Ultrasound Pulmonary Disease Quantification Using an Unsupervised Machine Learning Technique for COVID-19
COVID-19 is an ongoing global health pandemic. Although COVID-19 can be diagnosed with various tests such as PCR, these tests do not establish pulmonary disease burden. Whereas point-of-care lung ultrasound (POCUS) can directly assess the severity of characteristic pulmonary findings of COVID-19, the advantage of using US is that it is inexpensive, portable, and widely available for use in many clinical settings. For automated assessment of pulmonary findings, we have developed an unsupervised learning technique termed the calculated lung ultrasound (CLU) index. The CLU can quantify various types of lung findings, such as A or B lines, consolidations, and pleural effusions, and it uses these findings to calculate a CLU index score, which is a quantitative measure of pulmonary disease burden. This is accomplished using an unsupervised, patient-specific approach that does not require training on a large dataset. The CLU was tested on 52 lung ultrasound examinations from several institutions. CLU demonstrated excellent concordance with radiologist findings in different pulmonary disease states. Given the global nature of COVID-19, the CLU would be useful for sonographers and physicians in resource-strapped areas with limited ultrasound training and diagnostic capacities for more accurate assessment of pulmonary status.
- Open Access
- 19 December 2022
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 19  (24) , 17041
A Forehead Wearable Sensor for the Objective Measurement of Chronic Pain
Chronic pain impacts one in five Americans and is difficult to manage, costing ~USD 600 billion annually. The subjective experience of pain is a complex processing of central nervous system input. Recent advances in magnetic resonance imaging revealed the prefrontal cortex as vital to the perception of pain and that changes in the cerebral hemodynamics can be used to detect painful sensations. Current pain monitoring is dependent on the subjective rating provided by patients and is limited to a single time point. We have developed a biomarker for the objective, real-time and continuous chronic pain assessment using proprietary algorithms termed ROPA and cerebral optical spectrometry. Using a forehead sensor, the cerebral optical spectrometry data were collected in two clinical sites from 41 patients (19 and 22, respectively, from sites 1 and 2), who elected to receive an epidural steroid injection for the treatment of chronic pain. Patients rated their pain on a numeric rating scale, ranging from 0–10, which were used to validate the ROPA objective pain scoring. Multiple time points, including pre- and post-procedure were recorded. The steroid injection was performed per standard medical practice. There was a significant correlation between the patient’s reported numeric rating scale and ROPA, for both clinical sites (Overall ~0.81). Holding the subjective pain ratings on a numeric rating scale as ground truth, we determined that the area under the receiver operator curves for both sites revealed at least good (AUC: 64%) to excellent (AUC > 98%) distinctions between clinically meaningful pain severity differentiations (no/mild/moderate/severe). The objective measure of chronic pain (ROPA) determined using cerebral optical spectrometry significantly correlated with the subjective pain scores reported by the subjects. This technology may provide a useful method of detection for the objective and continuous monitoring and treatment of patients with chronic pain, particularly in clinical circumstances where direct assessment is not available, or to complement the patient-reported pain scores.
- 19 December 2022
- Published by MDPI AG in International Journal of Environmental Research and Public Health
- Vol. 19  (24) , 17041
Open AccessA Forehead Wearable Sensor for the Objective Measurement of Chronic Pain
Chronic pain impacts one in five Americans and is difficult to manage, costing ~USD 600 billion annually. The subjective experience of pain is a complex processing of central nervous system input. Recent advances in magnetic resonance imaging revealed the prefrontal cortex as vital to the perception of pain and that changes in the cerebral hemodynamics can be used to detect painful sensations. Current pain monitoring is dependent on the subjective rating provided by patients and is limited to a single time point. We have developed a biomarker for the objective, real-time and continuous chronic pain assessment using proprietary algorithms termed ROPA and cerebral optical spectrometry. Using a forehead sensor, the cerebral optical spectrometry data were collected in two clinical sites from 41 patients (19 and 22, respectively, from sites 1 and 2), who elected to receive an epidural steroid injection for the treatment of chronic pain. Patients rated their pain on a numeric rating scale, ranging from 0–10, which were used to validate the ROPA objective pain scoring. Multiple time points, including pre- and post-procedure were recorded. The steroid injection was performed per standard medical practice. There was a significant correlation between the patient’s reported numeric rating scale and ROPA, for both clinical sites (Overall ~0.81). Holding the subjective pain ratings on a numeric rating scale as ground truth, we determined that the area under the receiver operator curves for both sites revealed at least good (AUC: 64%) to excellent (AUC > 98%) distinctions between clinically meaningful pain severity differentiations (no/mild/moderate/severe). The objective measure of chronic pain (ROPA) determined using cerebral optical spectrometry significantly correlated with the subjective pain scores reported by the subjects. This technology may provide a useful method of detection for the objective and continuous monitoring and treatment of patients with chronic pain, particularly in clinical circumstances where direct assessment is not available, or to complement the patient-reported pain scores.
- Open Access
- 25 August 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12  (17) , 8490
Simulated Annealing-Based Wavelength Selection for Robust Tissue Oxygenation Estimation Powered by the Extended Modified Lambert–Beer Law
In this paper, we present a set of algorithms to enable the development of inexpensive hyperspectral sensors capable of estimating tissue oxygenation for wound monitoring. Estimation is conducted using the extended modified Lambert–Beer law, which has previously been proven robust to differences in melanin concentration. We introduce a novel wavelength selection algorithm that enables the estimation to be performed with high accuracy using only a small number (5–10) of wavelengths. Validation performed with Monte Carlo simulation data resulted in prediction errors <1%, with no significant differences among various skin types, for as few as five wavelengths under conditions representing both high precision instrumentation and more cost-effective sensors designed with inexpensive LEDs and/or filters. Validation with in vivo data collected from an occlusion study with 13 Asian volunteers showed statistically significant separation between the estimates for the at-rest and arterial occlusion states. Additional stability testing proved the proposed algorithms to be robust to small changes in the selected wavelengths as may occur in a real LED due to manufacturing tolerances and temperature fluctuations. This work concluded that the development of an inexpensive hyperspectral device for wound monitoring in all skin types is feasible using just a small number of wavelengths.
- 25 August 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12  (17) , 8490
Open AccessSimulated Annealing-Based Wavelength Selection for Robust Tissue Oxygenation Estimation Powered by the Extended Modified Lambert–Beer Law
In this paper, we present a set of algorithms to enable the development of inexpensive hyperspectral sensors capable of estimating tissue oxygenation for wound monitoring. Estimation is conducted using the extended modified Lambert–Beer law, which has previously been proven robust to differences in melanin concentration. We introduce a novel wavelength selection algorithm that enables the estimation to be performed with high accuracy using only a small number (5–10) of wavelengths. Validation performed with Monte Carlo simulation data resulted in prediction errors <1%, with no significant differences among various skin types, for as few as five wavelengths under conditions representing both high precision instrumentation and more cost-effective sensors designed with inexpensive LEDs and/or filters. Validation with in vivo data collected from an occlusion study with 13 Asian volunteers showed statistically significant separation between the estimates for the at-rest and arterial occlusion states. Additional stability testing proved the proposed algorithms to be robust to small changes in the selected wavelengths as may occur in a real LED due to manufacturing tolerances and temperature fluctuations. This work concluded that the development of an inexpensive hyperspectral device for wound monitoring in all skin types is feasible using just a small number of wavelengths.
- Open Access
- 26 January 2022
- Published by MDPI AG in Informatics
- Vol. 9  (1) , 10
Predictive Model for ICU Readmission Based on Discharge Summaries Using Machine Learning and Natural Language Processing
Predicting ICU readmission risk will help physicians make decisions regarding discharge. We used discharge summaries to predict ICU 30-day readmission risk using text mining and machine learning (ML) with data from the Medical Information Mart for Intensive Care III (MIMIC-III). We used Natural Language Processing (NLP) and the Bag-of-Words approach on discharge summaries to build a Document-Term-Matrix with 3000 features. We compared the performance of support vector machines with the radial basis function kernel (SVM-RBF), adaptive boosting (AdaBoost), quadratic discriminant analysis (QDA), least absolute shrinkage and selection operator (LASSO), and Ridge Regression. A total of 4000 patients were used for model training and 6000 were used for validation. Using the bag-of-words determined by NLP, the area under the receiver operating characteristic (AUROC) curve was 0.71, 0.68, 0.65, 0.69, and 0.65 correspondingly for SVM-RBF, AdaBoost, QDA, LASSO, and Ridge Regression. We then used the SVM-RBF model for feature selection by incrementally adding features to the model from 1 to 3000 bag-of-words. Through this exhaustive search approach, only 825 features (words) were dominant. Using those selected features, we trained and validated all ML models. The AUROC curve was 0.74, 0.69, 0.67, 0.70, and 0.71 respectively for SVM-RBF, AdaBoost, QDA, LASSO, and Ridge Regression. Overall, this technique could predict ICU readmission relatively well.
- 26 January 2022
- Published by MDPI AG in Informatics
- Vol. 9  (1) , 10
Open AccessPredictive Model for ICU Readmission Based on Discharge Summaries Using Machine Learning and Natural Language Processing
Predicting ICU readmission risk will help physicians make decisions regarding discharge. We used discharge summaries to predict ICU 30-day readmission risk using text mining and machine learning (ML) with data from the Medical Information Mart for Intensive Care III (MIMIC-III). We used Natural Language Processing (NLP) and the Bag-of-Words approach on discharge summaries to build a Document-Term-Matrix with 3000 features. We compared the performance of support vector machines with the radial basis function kernel (SVM-RBF), adaptive boosting (AdaBoost), quadratic discriminant analysis (QDA), least absolute shrinkage and selection operator (LASSO), and Ridge Regression. A total of 4000 patients were used for model training and 6000 were used for validation. Using the bag-of-words determined by NLP, the area under the receiver operating characteristic (AUROC) curve was 0.71, 0.68, 0.65, 0.69, and 0.65 correspondingly for SVM-RBF, AdaBoost, QDA, LASSO, and Ridge Regression. We then used the SVM-RBF model for feature selection by incrementally adding features to the model from 1 to 3000 bag-of-words. Through this exhaustive search approach, only 825 features (words) were dominant. Using those selected features, we trained and validated all ML models. The AUROC curve was 0.74, 0.69, 0.67, 0.70, and 0.71 respectively for SVM-RBF, AdaBoost, QDA, LASSO, and Ridge Regression. Overall, this technique could predict ICU readmission relatively well.
- Open Access
- 11 November 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11  (22) , 10628
Simulated Annealing-Based Hyperspectral Data Optimization for Fish Species Classification: Can the Number of Measured Wavelengths Be Reduced?
Relative to standard red/green/blue (RGB) imaging systems, hyperspectral imaging systems offer superior capabilities but tend to be expensive and complex, requiring either a mechanically complex push-broom line scanning method, a tunable filter, or a large set of light emitting diodes (LEDs) to collect images in multiple wavelengths. This paper proposes a new methodology to support the design of a hypothesized system that uses three imaging modes—fluorescence, visible/near-infrared (VNIR) reflectance, and shortwave infrared (SWIR) reflectance—to capture narrow-band spectral data at only three to seven narrow wavelengths. Simulated annealing is applied to identify the optimal wavelengths for sparse spectral measurement with a cost function based on the accuracy provided by a weighted k-nearest neighbors (WKNN) classifier, a common and relatively robust machine learning classifier. Two separate classification approaches are presented, the first using a multi-layer perceptron (MLP) artificial neural network trained on sparse data from the three individual spectra and the second using a fusion of the data from all three spectra. The results are compared with those from four alternative classifiers based on common machine learning algorithms. To validate the proposed methodology, reflectance and fluorescence spectra in these three spectroscopic modes were collected from fish fillets and used to classify the fillets by species. Accuracies determined from the two classification approaches are compared with benchmark values derived by training the classifiers with the full resolution spectral data. The results of the single-layer classification study show accuracies ranging from ~68% for SWIR reflectance to ~90% for fluorescence with just seven wavelengths. The results of the fusion classification study show accuracies of about 95% with seven wavelengths and more than 90% even with just three wavelengths. Reducing the number of required wavelengths facilitates the creation of rapid and cost-effective spectral imaging systems that can be used for widespread analysis in food monitoring/food fraud, agricultural, and biomedical applications.
- 11 November 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11  (22) , 10628
Open AccessSimulated Annealing-Based Hyperspectral Data Optimization for Fish Species Classification: Can the Number of Measured Wavelengths Be Reduced?
Relative to standard red/green/blue (RGB) imaging systems, hyperspectral imaging systems offer superior capabilities but tend to be expensive and complex, requiring either a mechanically complex push-broom line scanning method, a tunable filter, or a large set of light emitting diodes (LEDs) to collect images in multiple wavelengths. This paper proposes a new methodology to support the design of a hypothesized system that uses three imaging modes—fluorescence, visible/near-infrared (VNIR) reflectance, and shortwave infrared (SWIR) reflectance—to capture narrow-band spectral data at only three to seven narrow wavelengths. Simulated annealing is applied to identify the optimal wavelengths for sparse spectral measurement with a cost function based on the accuracy provided by a weighted k-nearest neighbors (WKNN) classifier, a common and relatively robust machine learning classifier. Two separate classification approaches are presented, the first using a multi-layer perceptron (MLP) artificial neural network trained on sparse data from the three individual spectra and the second using a fusion of the data from all three spectra. The results are compared with those from four alternative classifiers based on common machine learning algorithms. To validate the proposed methodology, reflectance and fluorescence spectra in these three spectroscopic modes were collected from fish fillets and used to classify the fillets by species. Accuracies determined from the two classification approaches are compared with benchmark values derived by training the classifiers with the full resolution spectral data. The results of the single-layer classification study show accuracies ranging from ~68% for SWIR reflectance to ~90% for fluorescence with just seven wavelengths. The results of the fusion classification study show accuracies of about 95% with seven wavelengths and more than 90% even with just three wavelengths. Reducing the number of required wavelengths facilitates the creation of rapid and cost-effective spectral imaging systems that can be used for widespread analysis in food monitoring/food fraud, agricultural, and biomedical applications.