Dataset Information

Automated segmentation of haematoma and perihaematomal oedema in MRI of acute spontaneous intracerebral haemorrhage.

ABSTRACT:

Background

Spontaneous intracerebral haemorrhage (SICH) is a common condition with high morbidity and mortality. Segmentation of haematoma and perihaematoma oedema on medical images provides quantitative outcome measures for clinical trials and may provide important markers of prognosis in people with SICH.

Methods

We take advantage of improved contrast seen on magnetic resonance (MR) images of patients with acute and early subacute SICH and introduce an automated algorithm for haematoma and oedema segmentation from these images. To our knowledge, there is no previously proposed segmentation technique for SICH that utilises MR images directly. The method is based on shape and intensity analysis for haematoma segmentation and voxel-wise dynamic thresholding of hyper-intensities for oedema segmentation.

Results

Using Dice scores to measure segmentation overlaps between labellings yielded by the proposed algorithm and five different expert raters on 18 patients, we observe that our technique achieves overlap scores that are very similar to those obtained by pairwise expert rater comparison. A further comparison between the proposed method and a state-of-the-art Deep Learning segmentation on a separate set of 32 manually annotated subjects confirms the proposed method can achieve comparable results with very mild computational burden and in a completely training-free and unsupervised way.

Conclusion

Our technique can be a computationally light and effective way to automatically delineate haematoma and oedema extent directly from MR images. Thus, with increasing use of MR images clinically after intracerebral haemorrhage this technique has the potential to inform clinical practice in the future.

SUBMITTER: Pszczolkowski S

PROVIDER: S-EPMC6382492 | biostudies-literature | 2019 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automated segmentation of haematoma and perihaematomal oedema in MRI of acute spontaneous intracerebral haemorrhage.

Pszczolkowski Stefan S Law Zhe K ZK Gallagher Rebecca G RG Meng Dewen D Swienton David J DJ Morgan Paul S PS Bath Philip M PM Sprigg Nikola N Dineen Rob A RA

Computers in biology and medicine 20190129

<h4>Background</h4>Spontaneous intracerebral haemorrhage (SICH) is a common condition with high morbidity and mortality. Segmentation of haematoma and perihaematoma oedema on medical images provides quantitative outcome measures for clinical trials and may provide important markers of prognosis in people with SICH.<h4>Methods</h4>We take advantage of improved contrast seen on magnetic resonance (MR) images of patients with acute and early subacute SICH and introduce an automated algorithm for ha ...[more]

PMID: 30711800

Similar Datasets

Project description:BackgroundOutcome after spontaneous (non-traumatic) intracerebral haemorrhage (ICH) is influenced by haematoma volume; up to one-third of ICHs enlarge within 24 hours of onset. Early haemostatic therapy might improve outcome by limiting haematoma growth. This is an update of a Cochrane Review first published in 2006, and last updated in 2009.ObjectivesTo examine 1) the effectiveness and safety of individual classes of haemostatic therapies, compared against placebo or open control, in adults with acute spontaneous intracerebral haemorrhage, and 2) the effects of each class of haemostatic therapy according to the type of antithrombotic drug taken immediately before ICH onset (i.e. anticoagulant, antiplatelet, or none).Search methodsWe searched the Cochrane Stroke Trials Register, CENTRAL; 2017, Issue 11, MEDLINE Ovid, and Embase Ovid on 27 November 2017. In an effort to identify further published, ongoing, and unpublished randomised controlled trials (RCT), we scanned bibliographies of relevant articles and searched international registers of RCTs in November 2017.Selection criteriaWe sought randomised controlled trials (RCTs) of any haemostatic intervention (i.e. pro-coagulant treatments such as coagulation factors, antifibrinolytic drugs, or platelet transfusion) for acute spontaneous ICH, compared with placebo, open control, or an active comparator, reporting relevant clinical outcome measures.Data collection and analysisTwo authors independently extracted data, assessed risk of bias, and contacted corresponding authors of eligible RCTs for specific data if they were not provided in the published report of an RCT.Main resultsWe included 12 RCTs involving 1732 participants. There were seven RCTs of blood clotting factors versus placebo or open control involving 1480 participants, three RCTs of antifibrinolytic drugs versus placebo or open control involving 57 participants, one RCT of platelet transfusion versus open control involving 190 participants, and one RCT of blood clotting factors versus fresh frozen plasma involving five participants. We were unable to include two eligible RCTs because they presented aggregate data for adults with ICH and other types of intracranial haemorrhage. We identified 10 ongoing RCTs. Across all seven criteria in the 12 included RCTs, the risk of bias was unclear in 37 (44%), high in 16 (19%), and low in 31 (37%). Only one RCT was at low risk of bias in all criteria.In one RCT of platelet transfusion versus open control for acute spontaneous ICH associated with antiplatelet drug use, there was a significant increase in death or dependence (modified Rankin Scale score 4 to 6) at day 90 (70/97 versus 52/93; risk ratio (RR) 1.29, 95% confidence interval (CI) 1.04 to 1.61, one trial, 190 participants, moderate-quality evidence). All findings were non-significant for blood clotting factors versus placebo or open control for acute spontaneous ICH with or without surgery (moderate-quality evidence), for antifibrinolytic drugs versus placebo (moderate-quality evidence) or open control for acute spontaneous ICH (moderate-quality evidence), and for clotting factors versus fresh frozen plasma for acute spontaneous ICH associated with anticoagulant drug use (no evidence).Authors' conclusionsBased on moderate-quality evidence from one trial, platelet transfusion seems hazardous in comparison to standard care for adults with antiplatelet-associated ICH.We were unable to draw firm conclusions about the efficacy and safety of blood clotting factors for acute spontaneous ICH with or without surgery, antifibrinolytic drugs for acute spontaneous ICH, and clotting factors versus fresh frozen plasma for acute spontaneous ICH associated with anticoagulant drug use.Further RCTs are warranted, and we await the results of the 10 ongoing RCTs with interest.

Project description:ObjectiveTo obtain precise estimates of age, haematoma volume, secondary haematoma expansion (HE) and mortality for patients with intracerebral haemorrhage (ICH) taking oral anticoagulants [Vitamin K antagonists (VKA-ICH) or non-Vitamin K antagonist oral anticoagulants (NOAC-ICH)] and those not taking oral anticoagulants (non-OAC ICH) at ICH symptom onset.MethodsWe conducted a systematic review and meta-analysis of studies comparing VKA-ICH or NOAC-ICH or both with non-OAC ICH. Primary outcomes were haematoma volume (in ml), HE, and mortality (in-hospital and 3-month). We calculated odds ratios (ORs) using the Mantel-Haenszel random-effects method and corresponding 95% confidence intervals (95%CI) and determined the mean ICH volume difference.ResultsWe identified 19 studies including data from 16,546 patients with VKA-ICH and 128,561 patients with non-OAC ICH. Only 2 studies reported data on 4943 patients with NOAC-ICH. Patients with VKA-ICH were significantly older than patients with non-OAC ICH (mean age difference: 5.55 years, 95%CI 4.03-7.07, p?<?0.0001, I2?=?92%, p?<?0.001). Haematoma volume was significantly larger in VKA-ICH with a mean difference of 9.66 ml (95%CI 6.24-13.07 ml, p?<?0.00001; I2?=?42%, p?=?0.05). HE occurred significantly more often in VKA-ICH (OR 2.96, 95%CI 1.74-4.97, p?<?0.00001; I2?=?65%). VKA-ICH was associated with significantly higher in-hospital mortality (VKA-ICH: 32.8% vs. non-OAC ICH: 22.4%; OR 1.83, 95%CI 1.61-2.07, p?<?0.00001, I2?=?20%, p?=?0.27) and 3-month mortality (VKA-ICH: 47.1% vs. non-OAC ICH: 25.5%; OR 2.24, 95%CI 1.52-3.31, p?<?0.00001, I2?=?71%, p?=?0.001). We did not find sufficient data for a meta-analysis comparing NOAC-ICH and non-OAC-ICH.ConclusionThis meta-analysis confirms, refines and expands findings from prior studies. We provide precise estimates of key prognostic factors and outcomes for VKA-ICH, which has larger haematoma volume, increased rate of HE and higher mortality compared to non-OAC ICH. There are insufficient data on NOACs.

Project description:ObjectivesTo test radiomics-based features extracted from noncontrast CT of patients with spontaneous intracerebral haemorrhage for prediction of haematoma expansion and poor functional outcome and compare them with radiological signs and clinical factors.Materials and methodsSeven hundred fifty-four radiomics-based features were extracted from 1732 scans derived from the TICH-2 multicentre clinical trial. Features were harmonised and a correlation-based feature selection was applied. Different elastic-net parameterisations were tested to assess the predictive performance of the selected radiomics-based features using grid optimisation. For comparison, the same procedure was run using radiological signs and clinical factors separately. Models trained with radiomics-based features combined with radiological signs or clinical factors were tested. Predictive performance was evaluated using the area under the receiver operating characteristic curve (AUC) score.ResultsThe optimal radiomics-based model showed an AUC of 0.693 for haematoma expansion and an AUC of 0.783 for poor functional outcome. Models with radiological signs alone yielded substantial reductions in sensitivity. Combining radiomics-based features and radiological signs did not provide any improvement over radiomics-based features alone. Models with clinical factors had similar performance compared to using radiomics-based features, albeit with low sensitivity for haematoma expansion. Performance of radiomics-based features was boosted by incorporating clinical factors, with time from onset to scan and age being the most important contributors for haematoma expansion and poor functional outcome prediction, respectively.ConclusionRadiomics-based features perform better than radiological signs and similarly to clinical factors on the prediction of haematoma expansion and poor functional outcome. Moreover, combining radiomics-based features with clinical factors improves their performance.Key points• Linear models based on CT radiomics-based features perform better than radiological signs on the prediction of haematoma expansion and poor functional outcome in the context of intracerebral haemorrhage. • Linear models based on CT radiomics-based features perform similarly to clinical factors known to be good predictors. However, combining these clinical factors with radiomics-based features increases their predictive performance.

Dataset Information

Automated segmentation of haematoma and perihaematomal oedema in MRI of acute spontaneous intracerebral haemorrhage.

Background

Methods

Results

Conclusion

Publications

Automated segmentation of haematoma and perihaematomal oedema in MRI of acute spontaneous intracerebral haemorrhage.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets