Deep Learning-Based Prediction of Hematoma Expansion Using a Single Brain Computed Tomographic Slice in Patients With Spontaneous Intracerebral Hemorrhages

Tang, Zhi-Ri; Zhu, Yiqin; Lu, Xin; Wu, Dengjun; Fan, Xinlin; Shen, Junjun; Chen, Gong; Zhao, Jian‐Lan; Xie, Rong; Xiao, Limin

doi:10.1016/j.wneu.2022.05.109

Cited by 13 publications

(13 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…S et al 2022/ USA [ 29 ] Prospective ICH DNN NA Single/Real-time data 2D 98 99 NA 0.99 Seyam, M., et al 2022/ Switzerland [ 40 ] Prospective ICH DL 256-section scanners (Somatom Force and Somatom Definition Flash, Siemens) Single/Real-time data 2D 87.2 93.9 93 NA Altuve, M., & Pérez, A. 2022/Venezuela [ 22 ] Retrospective ICH ResNet-18 NA Single/Real-time data 2D 95.65 96.2 95.93 NA Tang, Z., et al 2022/China[ 41 ] Retrospective ICH CNN NA Single/Real-time data 2D 91.97 88.37 90.58 NA Cortes-Ferre L, et al 2022/ Spain [ 26 ] Retrospective ICH DL NA Single/Benchmark 2D 91.4 94 92.7 0.978 Kau, T., et al 2022/ Austria [ 30 ] ...…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Diagnostic test accuracy of machine learning algorithms for the detection intracranial hemorrhage: a systematic review and meta-analysis study

Maghami,

Sattari,

Tahmasbi

et al. 2023

BioMed Eng OnLine

View full text Add to dashboard Cite

Background This systematic review and meta-analysis were conducted to objectively evaluate the evidence of machine learning (ML) in the patient diagnosis of Intracranial Hemorrhage (ICH) on computed tomography (CT) scans. Methods Until May 2023, systematic searches were conducted in ISI Web of Science, PubMed, Scopus, Cochrane Library, IEEE Xplore Digital Library, CINAHL, Science Direct, PROSPERO, and EMBASE for studies that evaluated the diagnostic precision of ML model-assisted ICH detection. Patients with and without ICH as the target condition who were receiving CT-Scan were eligible for the research, which used ML algorithms based on radiologists' reports as the gold reference standard. For meta-analysis, pooled sensitivities, specificities, and a summary receiver operating characteristics curve (SROC) were used. Results At last, after screening the title, abstract, and full paper, twenty-six retrospective and three prospective, and two retrospective/prospective studies were included. The overall (Diagnostic Test Accuracy) DTA of retrospective studies with a pooled sensitivity was 0.917 (95% CI 0.88–0.943, I2 = 99%). The pooled specificity was 0.945 (95% CI 0.918–0.964, I2 = 100%). The pooled diagnostic odds ratio (DOR) was 219.47 (95% CI 104.78–459.66, I2 = 100%). These results were significant for the specificity of the different network architecture models (p-value = 0.0289). However, the results for sensitivity (p-value = 0.6417) and DOR (p-value = 0.2187) were not significant. The ResNet algorithm has higher pooled specificity than other algorithms with 0.935 (95% CI 0.854–0.973, I2 = 93%). Conclusion This meta-analysis on DTA of ML algorithms for detecting ICH by assessing non-contrast CT-Scans shows the ML has an acceptable performance in diagnosing ICH. Using ResNet in ICH detection remains promising prediction was improved via training in an Architecture Learning Network (ALN).

show abstract

Section: Resultsmentioning

confidence: 99%

“…The Network Architecture analysis was divided into ResNet, RF, and SVM [ 20 – 26 , 28 , 30 – 39 , 41 – 43 , 45 , 46 , 48 – 50 ]. These results were significant for the specificity of the different network architecture models ( p -value = 0.0289).…”

Section: Resultsmentioning

confidence: 99%

Diagnostic test accuracy of machine learning algorithms for the detection intracranial hemorrhage: a systematic review and meta-analysis study

Maghami,

Sattari,

Tahmasbi

et al. 2023

BioMed Eng OnLine

View full text Add to dashboard Cite

show abstract

“…Semerano et al ( 30 ) also reported a higher rate of sICH among patients with reduced collaterals. Tang et al ( 31 , 32 ) used deep learning technology to create a model predictive of ICH expansion, which was derived from the baseline characteristics among the patients who suffered from ICH. The model had 90% accuracy and used the age of the patient, low-density lipoprotein cholesterol, time from onset to admission, systolic blood pressure, coagulopathy, baseline Glasgow Coma Scale, baseline NIHSS and the presence of intraventricular expansion to predict ICH expansion ( 31 ).…”

Section: Discussionmentioning

confidence: 99%

“…Tang et al ( 31 , 32 ) used deep learning technology to create a model predictive of ICH expansion, which was derived from the baseline characteristics among the patients who suffered from ICH. The model had 90% accuracy and used the age of the patient, low-density lipoprotein cholesterol, time from onset to admission, systolic blood pressure, coagulopathy, baseline Glasgow Coma Scale, baseline NIHSS and the presence of intraventricular expansion to predict ICH expansion ( 31 ). While the investigators examined the hematoma location, the authors did not evaluate occluded vessels, collateral circulation, or treatment with IAT as predictors ( 31 , 32 ).…”

Section: Discussionmentioning

confidence: 99%

“…The model had 90% accuracy and used the age of the patient, low-density lipoprotein cholesterol, time from onset to admission, systolic blood pressure, coagulopathy, baseline Glasgow Coma Scale, baseline NIHSS and the presence of intraventricular expansion to predict ICH expansion ( 31 ). While the investigators examined the hematoma location, the authors did not evaluate occluded vessels, collateral circulation, or treatment with IAT as predictors ( 31 , 32 ). Further investigation of these variables as predictors of ICH expansion is warranted.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Oxidation‑reduction potential parameters worsen following intraarterial therapy in patients with reduced collateral circulation and middle cerebral artery occlusions

et al. 2023

View full text Add to dashboard Cite

Collateral circulation is important for cerebral perfusion in acute ischemic strokes. Monitoring the oxidation-reduction potential (ORP) may be useful to assess collateral status or treatment efficacy. The objectives of the present study were to determine if the ORP was associated with collateral circulation status in middle cerebral artery (MCA) occlusions and to identify patterns in the ORP and the collateral circulation status among patients treated with intraarterial therapy (IAT) over time. The present pilot study was nested within a prospective cohort study measuring the ORP of the peripheral venous plasma of stroke patients. The population included in the present study were patients with MCA (M1/M2) occlusions. Two ORP parameters were examined: Static ORP (sORP; mV), indicating oxidative stress, and capacity ORP (cORP; µC), indicating antioxidant reserves. Collateral status was retrospectively graded using Miteff's system as good (grade 1) or reduced (grade 2/3). Comparisons were made between collateral status groups (reduced vs. good collaterals) in all patients, within a subset including only patients who received IAT, and between thrombolysis in cerebral infraction scale score (TICI) groups (0-2a vs. 2b/3). The Fisher's exact test, Student's t-test and Wilcoxon tests were used (α<0.20). The 19 patients were categorized based on their collaterals: Good collaterals (53%) and reduced collaterals (47%). The baseline characteristics were similar with the exception that the patients with good collaterals had a lower international normalized ratio (P=0.12) and were more likely to have a stroke on the left side (P=0.18) or to have a mismatch (P=0.05). The admission sORP values were comparable (169.5 vs. 164.2 mV; P=0.65), as was admission cORP (P=0.73). When considering only the patients who received IAT (n=12), admission sORP (P=0.69) and cORP (P=0.90) were also statistically similar. On day 2, after IAT, both groups experienced a worsening in ORP measures; however, the patients with good collaterals had a significantly lower sORP (169.4 vs. 203.5 mV; P=0.02) and a higher cORP (0.2 vs. 0.1 µC; P=0.002) compared with the patients with reduced collaterals. Neither sORP nor cORP were significantly different between TICI score groups on admission or on day 2. Upon discharge, patients with a TICI of 2b-3 had a significantly better sORP (P=0.03) and cORP (P=0.12) compared with those with a TICI of 0-2a. In conclusion, upon patient admission, the ORP parameters were not significantly different between the collateral circulation status groups for MCA occlusions. The ORP parameters worsened after IAT regardless of the collateral circulation status; however, after IAT, on day 2, patients with good collaterals experienced less oxidative stress (sORP) and had higher antioxidant reserves (cORP) than patients with reduced collaterals.

show abstract

Machine learning for predicting hematoma expansion in spontaneous intracerebral hemorrhage: a systematic review and meta-analysis

Liu,

Zhao,

Niu

et al. 2024

Neuroradiology

View full text Add to dashboard Cite

Deep Learning-Based Prediction of Hematoma Expansion Using a Single Brain Computed Tomographic Slice in Patients With Spontaneous Intracerebral Hemorrhages

Cited by 13 publications

References 34 publications

Diagnostic test accuracy of machine learning algorithms for the detection intracranial hemorrhage: a systematic review and meta-analysis study

Diagnostic test accuracy of machine learning algorithms for the detection intracranial hemorrhage: a systematic review and meta-analysis study

Oxidation‑reduction potential parameters worsen following intraarterial therapy in patients with reduced collateral circulation and middle cerebral artery occlusions

Machine learning for predicting hematoma expansion in spontaneous intracerebral hemorrhage: a systematic review and meta-analysis

Contact Info

Product

Resources

About