1
|
Zhao M, Li W, Hu Y, Jiang R, Zhao Y, Zhang D, Zhang Y, Wang R, Cao Y, Zhang Q, Ma Y, Li J, Yu S, Zhang R, Zheng Y, Wang S, Zhao J. Deep-learning tool for early identification of non-traumatic intracranial hemorrhage etiology and application in clinical diagnostics based on computed tomography (CT) scans. PeerJ 2025; 13:e18850. [PMID: 40028214 PMCID: PMC11871901 DOI: 10.7717/peerj.18850] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/17/2024] [Accepted: 12/20/2024] [Indexed: 03/05/2025] Open
Abstract
Background To develop an artificial intelligence system that can accurately identify acute non-traumatic intracranial hemorrhage (ICH) etiology (aneurysms, hypertensive hemorrhage, arteriovenous malformation (AVM), Moyamoya disease (MMD), cavernous malformation (CM), or other causes) based on non-contrast computed tomography (NCCT) scans and investigate whether clinicians can benefit from it in a diagnostic setting. Methods The deep learning model was developed with 1,868 eligible NCCT scans with non-traumatic ICH collected between January 2011 and April 2018. We tested the model on two independent datasets (TT200 and SD 98) collected after April 2018. The model's diagnostic performance was compared with clinicians' performance. We further designed a simulated study to compare the clinicians' performance with and without the deep learning system complements. Results The proposed deep learning system achieved area under the receiver operating curve of 0.986 (95% CI [0.967-1.000]) on aneurysms, 0.952 (0.917-0.987) on hypertensive hemorrhage, 0.950 (0.860-1.000) on arteriovenous malformation (AVM), 0.749 (0.586-0.912) on Moyamoya disease (MMD), 0.837 (0.704-0.969) on cavernous malformation (CM), and 0.839 (0.722-0.959) on other causes in TT200 dataset. Given a 90% specificity level, the sensitivities of our model were 97.1% and 90.9% for aneurysm and AVM diagnosis, respectively. On the test dataset SD98, the model achieved AUCs on aneurysms and hypertensive hemorrhage of 0.945 (95% CI [0.882-1.000]) and 0.883 (95% CI [0.818-0.948]), respectively. The clinicians achieve significant improvements in the sensitivity, specificity, and accuracy of diagnoses of certain hemorrhage etiologies with proposed system complements. Conclusions The proposed deep learning tool can be an accuracy tool for early identification of hemorrhage etiologies based on NCCT scans. It may also provide more information for clinicians for triage and further imaging examination selection.
Collapse
Affiliation(s)
- Meng Zhao
- Neurosurgery Department, Capital Medical University, Beijing, China
- China National Clinical Research Center for Neurological Diseases, Beijing, China
| | - Wenjie Li
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Yifan Hu
- Tencent You Tu Lab, Tencent, Shenzhen, China
| | - Ruixuan Jiang
- China National Clinical Research Center for Neurological Diseases, Beijing, China
| | - Yuanli Zhao
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Dong Zhang
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Yan Zhang
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Rong Wang
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Yong Cao
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Qian Zhang
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Yonggang Ma
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Jiaxi Li
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Shaochen Yu
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Ran Zhang
- Affiliated Hospital of Shandong Jining Medical College, Jining, Shandong, China
| | | | - Shuo Wang
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| | - Jizong Zhao
- China National Clinical Research Center for Neurological Diseases, Beijing, China
- Beijing Tiantan Hospital, Capital Medical University, Beijing, China
| |
Collapse
|
2
|
Lieberman OJ, Berkowitz AL. Diagnostic Approach to the Patient with Altered Mental Status. Semin Neurol 2024; 44:579-605. [PMID: 39353612 DOI: 10.1055/s-0044-1791245] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/04/2024]
Abstract
Acute encephalopathy is a common presenting symptom in the emergency room and complicates many hospital and intensive care unit admissions. The evaluation of patients with encephalopathy poses several challenges: limited history and examination due to the patient's mental status, broad differential diagnosis of systemic and neurologic etiologies, low yield of neurodiagnostic testing due to the high base rate of systemic causes, and the importance of identifying less common neurologic causes of encephalopathy that can be life-threatening if not identified and treated. This article discusses the differential diagnosis of acute encephalopathy, presents an approach to the history and examination in a patient with encephalopathy, reviews the literature on the yield of neurodiagnostic testing in this population, and provides a diagnostic framework for the evaluation of patients with altered mental status.
Collapse
|
3
|
Tan S, Tang C, Ng JS, Ng C, Kovoor J, Gupta A, Goh R, Bacchi S, Jannes J, Kleinig T. Delays in the diagnosis of ischaemic stroke presenting with persistent reduced level of consciousness: A systematic review. J Clin Neurosci 2023; 115:14-19. [PMID: 37454440 DOI: 10.1016/j.jocn.2023.07.009] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/12/2023] [Revised: 06/12/2023] [Accepted: 07/10/2023] [Indexed: 07/18/2023]
Abstract
INTRODUCTION Stroke presenting with a reduced level of consciousness (RLOC) may result in diagnostic error and/or delay. Missed or delayed diagnosis of acute ischaemic stroke may preclude otherwise applicable hyperacute stroke interventions. The frequency, reasons for, and consequences of diagnostic error and delay due to RLOC are uncertain. METHOD The databases PubMed, EMBASE, and Cochrane library were searched in adherence with the PRISMA guidelines. The systematic review was prospectively registered on PROSPERO. RESULTS Initial searches returned 1162 results, of which 6 fulfilled inclusion criteria. The majority of identified studies show that ischaemic stroke presenting with RLOC is at increased risk of missed or delayed diagnosis. Hyperacute stroke interventions may also be delayed. There is limited evidence regarding the reason for these delays; however, the delays may result from neuroimaging delay associated with diagnostic uncertainty. There is also limited evidence regarding the outcomes of patients with stroke and RLOC who experience diagnostic delay; however, the available literature suggests that outcomes may be poor, including motor and cognitive impairment, as well as long-term impaired consciousness. The included studies did not evaluate, but have suggested urgent MRI access, educational interventions, and protocolisation of the evaluation of RLOC as means to reduce poor outcomes. CONCLUSIONS Ischaemic stroke patients with RLOC are at risk of diagnostic delay and error. These patients may have poor outcomes. Additional research is required to identify the contributing factors more clearly and to provide amelioration strategies.
Collapse
Affiliation(s)
- Sheryn Tan
- University of Adelaide, Adelaide, SA 5005, Australia.
| | - Charis Tang
- University of Adelaide, Adelaide, SA 5005, Australia
| | - Jeng Swen Ng
- University of Adelaide, Adelaide, SA 5005, Australia
| | - Cleo Ng
- University of Adelaide, Adelaide, SA 5005, Australia
| | - Joshua Kovoor
- University of Adelaide, Adelaide, SA 5005, Australia; Royal Adelaide Hospital, Adelaide, SA 5000, Australia
| | - Aashray Gupta
- University of Adelaide, Adelaide, SA 5005, Australia; Gold Coast University Hospital, Southport, QLD 4215, Australia
| | - Rudy Goh
- University of Adelaide, Adelaide, SA 5005, Australia; Royal Adelaide Hospital, Adelaide, SA 5000, Australia; Lyell McEwin Hospital, Elizabeth Vale, SA 5112, Australia
| | - Stephen Bacchi
- University of Adelaide, Adelaide, SA 5005, Australia; Royal Adelaide Hospital, Adelaide, SA 5000, Australia; Flinders University, Bedford Park, SA 5042, Australia
| | - Jim Jannes
- University of Adelaide, Adelaide, SA 5005, Australia; Royal Adelaide Hospital, Adelaide, SA 5000, Australia
| | - Timothy Kleinig
- University of Adelaide, Adelaide, SA 5005, Australia; Royal Adelaide Hospital, Adelaide, SA 5000, Australia
| |
Collapse
|
4
|
Hill M, Moreda M, Navarro J, Mulkey M. Assessing Patients With Altered Level of Consciousness. Crit Care Nurse 2023; 43:58-65. [PMID: 37524369 PMCID: PMC10403291 DOI: 10.4037/ccn2023449] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 08/02/2023]
Abstract
Patients with alterations in level of consciousness are among the most difficult to assess, so knowledge of how to assess these patients is important for tracking trends and identifying changes. This article discusses methods used to assess patients admitted with an altered level of consciousness and describes the neurological assessment of and potential causes for altered level of consciousness. Identifying and understanding certain examination findings enable faster recognition and intervention for life-threatening neurological events, directly impacting outcomes for neurologically compromised individuals.
Collapse
Affiliation(s)
- Michelle Hill
- Michelle Hill is the Comprehensive Stroke Program coordinator, OhioHealth Riverside Methodist Hospital, Columbus, Ohio
| | - Melissa Moreda
- Melissa Moreda is an inpatient diabetes clinical nurse specialist, Duke Raleigh Hospital, Raleigh, North Carolina
| | - Jacqueline Navarro
- Jacqueline Navarro is an advanced practice clinician, pulmonary critical care, University of Utah Hospital, Salt Lake City, Utah
| | - Malissa Mulkey
- Malissa Mulkey is a postdoctoral research fellow, Indiana University-Purdue University Indianapolis, Indiana, and an intensive care unit clinical nurse specialist, University of North Carolina Rex Hospital, Raleigh
| |
Collapse
|
5
|
The diagnostic value of the neurological examination in coma of unknown etiology. J Neurol 2021; 268:3826-3834. [PMID: 33796895 PMCID: PMC8463407 DOI: 10.1007/s00415-021-10527-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/04/2020] [Revised: 03/18/2021] [Accepted: 03/22/2021] [Indexed: 11/20/2022]
Abstract
Background Identifying the cause of non-traumatic coma in the emergency department is challenging. The clinical neurological examination is the most readily available tool to detect focal neurological deficits as indicators for cerebral causes of coma. Previously proposed clinical pathways have granted the interpretation of clinical findings a pivotal role in the diagnostic work-up. We aimed to identify the actual diagnostic reliability of the neurological examination with regard to identifying acute brain damage.
Methods Eight hundred and fifty-three patients with coma of unknown etiology (CUE) were examined neurologically in the emergency department following a predefined routine. Coma-explaining pathologies were identified retrospectively and grouped into primary brain pathology with proof of acute brain damage and other causes without proof of acute structural pathology. Sensitivity, specificity and percentage of correct predictions of different examination protocols were calculated using contingency tables and binary logistic regression models. Results The full neurological examination was 74% sensitive and 60% specific to detect acute structural brain damage underlying CUE. Sensitivity and specificity were higher in non-sedated patients (87/61%) compared to sedated patients (64%/59%). A shortened four-item examination protocol focusing on pupils, gaze and pyramidal tract signs was only slightly less sensitive (67%) and more specific (65%).
Conclusions Due to limited diagnostic reliability of the physical examination, the absence of focal neurological signs in acutely comatose patients should not defer from a complete work-up including brain imaging. In an emergency, a concise neurological examination should thus serve as one part of a multimodal diagnostic approach to CUE.
Collapse
|
6
|
Lutz M, Möckel M, Lindner T, Ploner CJ, Braun M, Schmidt WU. The accuracy of initial diagnoses in coma: an observational study in 835 patients with non-traumatic disorder of consciousness. Scand J Trauma Resusc Emerg Med 2021; 29:15. [PMID: 33436034 PMCID: PMC7805149 DOI: 10.1186/s13049-020-00822-w] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/22/2020] [Accepted: 12/13/2020] [Indexed: 11/10/2022] Open
Abstract
BACKGROUND Management of patients with coma of unknown etiology (CUE) is a major challenge in most emergency departments (EDs). CUE is associated with a high mortality and a wide variety of pathologies that require differential therapies. A suspected diagnosis issued by pre-hospital emergency care providers often drives the first approach to these patients. We aim to determine the accuracy and value of the initial diagnostic hypothesis in patients with CUE. METHODS Consecutive ED patients presenting with CUE were prospectively enrolled. We obtained the suspected diagnoses or working hypotheses from standardized reports given by prehospital emergency care providers, both paramedics and emergency physicians. Suspected and final diagnoses were classified into I) acute primary brain lesions, II) primary brain pathologies without acute lesions and III) pathologies that affected the brain secondarily. We compared suspected and final diagnosis with percent agreement and Cohen's Kappa including sub-group analyses for paramedics and physicians. Furthermore, we tested the value of suspected and final diagnoses as predictors for mortality with binary logistic regression models. RESULTS Overall, suspected and final diagnoses matched in 62% of 835 enrolled patients. Cohen's Kappa showed a value of κ = .415 (95% CI .361-.469, p < .005). There was no relevant difference in diagnostic accuracy between paramedics and physicians. Suspected diagnoses did not significantly interact with in-hospital mortality (e.g., suspected class I: OR .982, 95% CI .518-1.836) while final diagnoses interacted strongly (e.g., final class I: OR 5.425, 95% CI 3.409-8.633). CONCLUSION In cases of CUE, the suspected diagnosis is unreliable, regardless of different pre-hospital care providers' qualifications. It is not an appropriate decision-making tool as it neither sufficiently predicts the final diagnosis nor detects the especially critical comatose patient. To avoid the risk of mistriage and unnecessarily delayed therapy, we advocate for a standardized diagnostic work-up for all CUE patients that should be triggered by the emergency symptom alone and not by any suspected diagnosis.
Collapse
Affiliation(s)
- Maximilian Lutz
- Department of Neurology, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Augustenburger Platz 1, 13353, Berlin, Germany
| | - Martin Möckel
- Department of Emergency Medicine, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Berlin, Germany
| | - Tobias Lindner
- Department of Emergency Medicine, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Berlin, Germany
| | - Christoph J Ploner
- Department of Neurology, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Augustenburger Platz 1, 13353, Berlin, Germany
| | - Mischa Braun
- Department of Neurology, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Augustenburger Platz 1, 13353, Berlin, Germany.,Center for Stroke Research, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Berlin, Germany
| | - Wolf Ulrich Schmidt
- Department of Neurology, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Augustenburger Platz 1, 13353, Berlin, Germany. .,Center for Stroke Research, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Berlin, Germany.
| |
Collapse
|