Charting the potential of brain computed tomography deep learning systems

Buchlak, Quinlan D.; Milne, Michael R.; Seah, Jarrel; Johnson, Andrew; Samarasinghe, Gihan; Hachey, Ben; Esmaili, Nazanin; Tran, Aengus; Leveque, Jean-Christophe; Farrokhi, Farrokh; Goldschlager, Tony; Edelstein, Simon; Brotchie, Peter

doi:10.1016/j.jocn.2022.03.014

Cited by 16 publications

(6 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast to mature imaging modalities, for example, magnetic resonance imaging (MRI) [1], computed tomography (CT) [2], positron emission tomography [3], and ultrasound (US) imaging [4], optical imaging uses a nonionizing radiation to obtain images noninvasively with high spatiotemporal resolution and sensitivity [5]. However, in the optical imaging of biological targets, light is easily dispersed and absorbed by endogenous biological tissues, including blood, fat, and skin, making it challenging to image deep tissues [6].…”

Section: Introductionmentioning

confidence: 99%

“…Finally, the challenges and perspectives of stimuliresponsive composite NIR-II probes are discussed. 2 | ONOO − -RESPONSIVE NIR-II IMAGING ONOO − is an RNS and a biomarker of the liver status because ONOO − overexpression induced by herbal medicines can cause hepatic damage [41,42]. In ONOO −responsive optical probes, strong oxidation by ONOO − can block the ACIE effect between organic species and downconversion nanoparticles (DCNPs).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Bioresponsive fluorescent probes active in the second near‐infrared window

Chen

Chang

2023

iRADIOLOGY

View full text Add to dashboard Cite

Emerging stimuli-responsive composite probes active in the second nearinfrared window (NIR-II, 1000-1700 nm) hold vast potential for improving in vivo imaging performance with minimized noise interference. The interactions among external irradiation, shell species, and the emissive core are key factors in the design of smart structures. The external irradiation provides energy for shell species and the emissive core to generate intense NIR-II fluorescence signals, while the energy transfer process hinders NIR-II emission in the inner structure of smart composite probes. However, if pathophysiological stimuli interrupt the above processes, then NIR-II fluorescence signals are recovered. This review covers NIR-II imaging based on diverse smart composite NIR-II fluorescent probes responding to various biological stimuli, including ONOO − , overexpressed reactive sulfur species, abnormally expressed enzymes, and abnormal levels of physiological metabolites. Finally, to appreciate these advances, the challenges and perspectives of stimuliresponsive composite NIR-II probes are highlighted.

show abstract

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

Bioresponsive fluorescent probes active in the second near‐infrared window

Chen

Chang

2023

iRADIOLOGY

View full text Add to dashboard Cite

show abstract

“…19 ML encompasses a range of functions performed by computer systems attempting to reproduce or exceed human-level cognitive capabilities using sophisticated, nonlinear algorithms. 20 In supervised ML tasks, large sets of labeled data are used to train algorithms that can classify or predict outcomes accurately and facilitate improved decision-making. 21 ML applications have been used in a variety of health care fields.…”

Section: Introductionmentioning

confidence: 99%

“…Machine learning (ML) is a subdiscipline of artificial intelligence (AI) bringing together mathematics, statistics, and computer science to create algorithms that learn from data and perform cognitive tasks 19 . ML encompasses a range of functions performed by computer systems attempting to reproduce or exceed human‐level cognitive capabilities using sophisticated, nonlinear algorithms 20 . In supervised ML tasks, large sets of labeled data are used to train algorithms that can classify or predict outcomes accurately and facilitate improved decision‐making 21 …”

Section: Introductionmentioning

confidence: 99%

Predicting Acoustic Hearing Preservation Following Cochlear Implant Surgery Using Machine Learning

et al. 2023

Self Cite

View full text Add to dashboard Cite

ObjectivesThe aim of the study was to train and test supervised machine‐learning classifiers to predict acoustic hearing preservation after CI using preoperative clinical data.Study DesignRetrospective predictive modeling study of prospectively collected single‐institution CI dataset.MethodsOne hundred and seventy‐five patients from a REDCap database including 761 patients >18 years who underwent CI and had audiometric testing preoperatively and one month after surgery were included. The primary outcome variable was the lowest quartile change in acoustic hearing at one month after CI using various formulae (standard pure tone average, SPTA; low‐frequency PTA, LFPTA). Analysis involved applying multivariate logistic regression to detect statistical associations and training and testing supervised learning classifiers. Classifier performance was assessed with numerous metrics including area under the receiver operating characteristic curve (AUC) and Matthews correlation coefficient (MCC).ResultsLowest quartile change (indicating hearing preservation) in SPTA was positively associated with a history of meningitis, preoperative LFPTA, and preoperative SPTA. Lowest quartile change in SPTA was negatively associated with sudden hearing loss, noise exposure, aural fullness, and abnormal anatomy. Lowest quartile change in LFPTA was positively associated with preoperative LFPTA. Lowest quartile change in LFPTA was negatively associated with tobacco use. Random forest demonstrated the highest mean classification performance on the validation dataset when predicting each of the outcome variables.ConclusionsMachine learning demonstrated utility for predicting preservation of residual acoustic hearing in patients undergoing CI surgery, and the detected associations facilitated the interpretation of our machine‐learning models. The models and statistical associations together may be used to facilitate improvements in shared clinical decision‐making and patient outcomes.Level of Evidence3 Laryngoscope, 2023

show abstract

“…Timely retrieval of the necessary information obtained by radiological examinations and its subsequent analysis by means of machine learning algorithms may facilitate fast and effective decisions in the diagnosis of different pathologies and improve the quality of the appropriate medical aid. It is especially important in the sphere of emergency care, for instance, in timely diagnosis of intracranial hemorrhages (ICH) [3][4][5].…”

Section: Introductionmentioning

confidence: 99%

Text Analysis of Radiology Reports with Signs of Intracranial Hemorrhage on Brain CT Scans Using the Decision Tree Algorithm

Khoruzhaya¹,

Kozlov²,

Arzamasov³

et al. 2022

Sovrem Tehnol Med

View full text Add to dashboard Cite

The aim of the study is to create, train, and test the algorithm for the analysis of brain CT text reports using a decision tree model to solve the task of simple binary classification of presence/absence of intracranial hemorrhage (ICH) signs.Materials and Methods. The initial data is a download from the Unified Radiological Information Service of the Unified Medical Information and Analytical System (URIS UMIAS) containing 34,188 studies obtained by a non-contrast CT of the brain in 56 inpatient medical settings. Data analysis and preprocessing were carried out using NLTK (Natural Language Toolkit, version 3.6.5), a library for symbolic and statistical processing of natural language, and scikit-learn, a machine learning library containing tools for classification tasks. According to 14 selected ICH-related key words, as well as 33 stop-phrases with key words denoting absence of ICH, an automatic selection of the CT investigations and their subsequent expert verification were carried out. Two classes of investigations were formed based on the sample from 3980 protocol descriptions: containing descriptions of ICH and without them. The problem of binary classification was solved using the decision tree algorithm as a model. To evaluate the performance of the model, the CT investigations were divided randomly into samples in the ratio of 7:3. Of 3980 protocols, 2786 were assigned to the training data set, 1194 -to the test one.Results. According to the test results, the designed and trained algorithm in the binary classification of the CT reports "with signs of ICH" and "without signs of ICH" has shown sensitivity of 0.94, specificity of 0.88, F-score of 0.83. Conclusion.The developed and trained algorithm for the analysis of radiology reports has demonstrated high accuracy in relation to brain CT with signs of intracranial hemorrhage and can be used to solve binary classification problems and create appropriate data sets. However, it is limited by the need for manual revision of CT studies to ensure quality control.

show abstract

Charting the potential of brain computed tomography deep learning systems

Cited by 16 publications

References 64 publications

Bioresponsive fluorescent probes active in the second near‐infrared window

Bioresponsive fluorescent probes active in the second near‐infrared window

Predicting Acoustic Hearing Preservation Following Cochlear Implant Surgery Using Machine Learning

Text Analysis of Radiology Reports with Signs of Intracranial Hemorrhage on Brain CT Scans Using the Decision Tree Algorithm

Contact Info

Product

Resources

About