Cases Of Intracranial Hemorrhage Research Articles

Revisiting the pathophysiology of intracranial hemorrhage in fetuses with Chiari II malformation - novel imaging biomarkers of disease severity?

Intracranial hemorrhage (ICH) has emerged as a notable concern in Chiari II malformation (CM II), yet its origins and clinical implications remain elusive. This study aims to validate the in-utero prevalence of ICH in CM II and investigate contributing factors, and visualize the findings in a network format. A single-center retrospective review of fetal MRI scans obtained in fetuses with CM II presenting (January 2007 to December 2022) was performed for ICH utilizing EPI-T2* blood-sensitive sequence. Fetuses with aqueduct stenosis (AS) were included as a control group. The incidence of ICH and corresponding gestational ages were compared between CM II and AS cases, and morphometric measurements (inner/outer CSF spaces, posterior fossa, venous structure) were compared among the four 1:1 age-matched groups: CM II+ICH, CM II-ICH, AS+ICH, and AS-ICH. Additionally, a co-occurrence network was constructed to visualize associations between phenotypic features in ICH cases. A total of 101 fetuses with CM II and 90 controls with AS at a median gestational age of 24.4 weeks and 22.8 weeks (P=.138) were included. Prevalence of ICH in fetuses with CM II was higher compared to the AS cases (28.7% vs 18.9%, P=.023), accompanied by congested veins (deep vein congestion mainly in young fetuses, and cortical veins may also be affected in older fetuses). ICH was notably correlated with specific anatomical features, essentially characterized by reduced outer cerebrospinal fluid spaces and clivus-supraocciput angle. The co-occurrence network analysis reveals complex connections including bony defects, small posterior fossa dimensions, vermis ectopia, reduced CSF spaces as well as venous congestion and venous sinus stenosis as pivotal components within the network. The high prevalence of ICH - detected by fetal MRI -among fetuses with CM emphasizes the pathophysiological importance of venous congestion, ICH, and vasogenic edema. As indicators of disease severity, these features may serve as helpful additional imaging biomarkers for the identification of potential candidates for fetal surgery.ABBREVIATIONS: CM II＝Chiari type II malformation; AS＝aqueduct stenosis; ICH ＝Intracranial hemorrhage.

Classification of Intracranial Hemorrhage (CT) Images Using CNN- LSTM Method and Image-Based GLCM Features

In this study, a hybrid approach is utilized, which combines image-based grey-level co-occurrence matrix attributes and transformed-based attributes. The combination of attribute-based techniques is more accurate in categorizing cerebral hemorrhages in CT images when compared with the transformed attribute-based and image-attribute-based approaches. Deep learning methods, particularly long short-term memory (LSTM), have gained popularity in natural language processing applications like text and sentiment analysis. This paper presents an in-depth deep-learning system that is fully automated. The purpose of the system is to organize radiological data and identify cases of intracranial hemorrhage (ICH) for diagnosis. The structure of the proposed DL model includes 1D convolution neural networks (CNN), long short-term memory (LSTM) units, and a logistic function. A significant dataset consisting of 12,852 radiological reports of head computed tomography (CT) was used to train and assess these components.

AI assisted reader evaluation in acute CT head interpretation (AI-REACT): protocol for a multireader multicase study

IntroductionA non-contrast CT head scan (NCCTH) is the most common cross-sectional imaging investigation requested in the emergency department. Advances in computer vision have led to development of several artificial intelligence...

Strengthening deep-learning models for intracranial hemorrhage detection: strongly annotated computed tomography images and model ensembles.

Multiple attempts at intracranial hemorrhage (ICH) detection using deep-learning techniques have been plagued by clinical failures. We aimed to compare the performance of a deep-learning algorithm for ICH detection trained on strongly and weakly annotated datasets, and to assess whether a weighted ensemble model that integrates separate models trained using datasets with different ICH improves performance. We used brain CT scans from the Radiological Society of North America (27,861 CT scans, 3,528 ICHs) and AI-Hub (53,045 CT scans, 7,013 ICHs) for training. DenseNet121, InceptionResNetV2, MobileNetV2, and VGG19 were trained on strongly and weakly annotated datasets and compared using independent external test datasets. We then developed a weighted ensemble model combining separate models trained on all ICH, subdural hemorrhage (SDH), subarachnoid hemorrhage (SAH), and small-lesion ICH cases. The final weighted ensemble model was compared to four well-known deep-learning models. After external testing, six neurologists reviewed 91 ICH cases difficult for AI and humans. InceptionResNetV2, MobileNetV2, and VGG19 models outperformed when trained on strongly annotated datasets. A weighted ensemble model combining models trained on SDH, SAH, and small-lesion ICH had a higher AUC, compared with a model trained on all ICH cases only. This model outperformed four deep-learning models (AUC [95% C.I.]: Ensemble model, 0.953[0.938-0.965]; InceptionResNetV2, 0.852[0.828-0.873]; DenseNet121, 0.875[0.852-0.895]; VGG19, 0.796[0.770-0.821]; MobileNetV2, 0.650[0.620-0.680]; p < 0.0001). In addition, the case review showed that a better understanding and management of difficult cases may facilitate clinical use of ICH detection algorithms. We propose a weighted ensemble model for ICH detection, trained on large-scale, strongly annotated CT scans, as no model can capture all aspects of complex tasks.

Examination of the Cerebellomedullary Cistern Using Postmortem Computed Tomography in Various Types of Intracranial Hemorrhage

Intracranial hemorrhage is a major cause of sudden unexpected death and its identification is important for death investigations. Cisternal puncture of the cerebellomedullary cistern (CMC) can be used to identify intracranial hemorrhage during postmortem examination. Intracranial hemorrhage comprises various types of hemorrhage, and the possibility of hemorrhage identification by cisternal puncture can differ according to the type of intracranial hemorrhage. Postmortem computed tomography (PMCT) is non-invasive and can be performed before autopsy. In this study, we aimed to identify the hemorrhage in the CMC using PMCT in cases of various intracranial hemorrhage. PMCT was performed before the autopsy. Autopsy reports and PMCT were retrospectively reviewed for 108 cases of intracranial hemorrhage confirmed by conventional autopsy. Hemorrhagic regions showed ≥60 Hounsfield units on PMCT. Hemorrhage in the CMC was identified in 42.6% (46/108) by PMCT and was frequently identified in the cases of basal subarachnoid hemorrhage (SAH) (19/21). Hemorrhage in the CMC was identified in approximately 50% of patients with non-basal SAH and intracerebral hemorrhage. Detection of hemorrhage in the CMC by PMCT is hindered by several factors, such as dental artifacts. This study does not guarantee the usefulness of postmortem cisternal puncture. However, this study verified hemorrhage in the CMC according to the various types of intracranial hemorrhages using PMCT and showed its possibilities and limitations.

First-in-human trial of a self-expandable, temporary dilation system for intracranial atherosclerotic disease in patients presenting with acute ischemic stroke

BackgroundIntracranial atherosclerotic disease (ICAD) significantly contributes to ischemic stroke, especially among Asian populations. Large vessel occlusion (LVO) due to underlying ICAD accounts for 15–35% of acute ischemic stroke cases requiring...

Abnormalities of brain imaging in COVID-19 patients with neurological symptoms.

Background: Coronavirus disease 2019 (COVID-19) is a multisystem disease, manifested by several symptoms of various degrees. Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) can affect the central nervous system (CNS) through several mechanisms and brain imaging plays an essential role in the diagnosis and evaluation of the neurological involvement of COVID-19. Moreover, brain imaging of patients with COVID-19 would result in a better understanding of SARS-CoV-2 neuro-pathophysiology. In this study, we evaluated the brain imaging findings of patients with COVID-19 in Shohada-e Tajrish Hospital, Tehran, Iran. Methods: This was a single-center, retrospective, and observational study. The hospital records and chest and brain computed tomography (CT) scans of patients with confirmed COVID-19 were reviewed. Results: 161 patients were included in this study (39.1% women, mean age: 60.84). Thirteen patients (8%) had ischemic strokes identified by brain CT. Subdural hematoma, subdural effusion, and subarachnoid hemorrhage were confirmed in three patients. Furthermore, there were four cases of intracranial hemorrhage (ICH) and intraventricular hemorrhage (IVH). Patients with and without abnormal brain CTs had similar average ages. The rate of brain CT abnormalities in both genders did not differ significantly. Moreover, abnormal brain CT was not associated with increased death rate. There was no significant difference in lung involvement (according to lung CT scan) between the two groups. Conclusion: Our experience revealed a wide range of imaging findings in patients with COVID-19 and these findings were not associated with a more severe lung involvement or increased rate of mortality.

Gilteritinib and the risk of intracranial hemorrhage: a case series of a possible, under-reported side effect

Gilteritinib is currently approved for patients with relapsed/refractory AML with FLT3 mutations, based on the positive results of the pivotal ADMIRAL study. In ADMIRAL trial, no increased risk of bleeding was reported, but in the previous dose finding study, a single event of intracranial hemorrhage (ICH) was registered after exposure to subtherapeutic doses of gilteritinib. Here, we report the first case series on five ICHs diagnosed in patients with FLT3-mutated AML, occurred within the first month of exposure to gilteritinib. Our cohort included 24 patients treated in three Italian centers. Most of these ICH cases were non-severe and self-limiting, while one was fatal. This link with ICHs remains in any case uncertain for the presence of active AML. We further reported that an analysis of the post-marketing surveillance data (EudraVigilance) retrieved other 11 cases of ICHs present in the database after gilteritinib treatment. A causality assessment was performed according to the Dx3 method to evaluate the possibility that ICHs might be an actual side effect of gilteritinib. In conclusion, further research is needed to elucidate the potential role of gilteritinib in the pathogenesis of ICHs.

Injuries caused by defensive bullet launchers and resource utilization during the French yellow vests protests: A retrospective study

BackgroundIn 2018, the French “Yellow Vest” social protest movement spread with weekly demonstrations resulting in confrontations between protesters and law enforcement. Non-lethal weapons, such as defensive bullet launchers (DBL) were used, and significant injuries have been reported through media, leading to public controversy regarding their use. These injuries are not well-known to civilian emergency physicians.The aim of this study is to describe the injuries caused by DBL among Emergency Department (ED) patients during these demonstrations and to identify the characteristics that required specialized care and hospital admission. MethodsA multicenter retrospective study was conducted in 7 EDs of academic hospitals in Paris, France. Adult ED patients who presented with DBL injuries during “yellow vest” strikes between November 2018 and May 2019 were included. The primary outcome was the rate of DBL patients requiring hospital admission. We also compared the characteristics of the injuries and the care provided between the admitted patients and other DBL patients. Results152 patients were included. 17% were admitted to hospital, with 19% of them being transferred to intensive care units. 49% of all patients had head, face, eye or neck injuries including 4 cases of intracranial hemorrhage, 1 carotide dissection, 1 laryngeal edema, 1 pneumencephalus. 11% of all patients presented with multiple wounds, and 28% had fractures (77% of admitted patients vs 18%, p < 0.001). Surgery was required for 20% of all patients (62% of admitted patients vs 10%, p < 0.001). Maxillofacial surgery was performed on 38% of admitted patients, orthopedic surgery on 25%, and neurosurgery on 13%. No death were reported. ConclusionThe use of DBL during the “yellow vest” civil strikes was associated with a high rate of head, face, eye or neck injuries among injured ED patients. Hospital admission was associated with a higher rate of fractures, with most of them requiring maxillofacial, orthopedic and neurosurgeries.

Endovascular Treatment and Peri-interventional Management of Ruptured Cerebrovascular Lesions During Pregnancy

PurposeHemorrhagic stroke, particularly occurring from ruptured cerebrovascular malformations, is responsible for 5–12% of all maternal deaths during pregnancy and the puerperium. Whether endovascular treatment is feasible and safe for both the mother and the fetus, is still a matter of debate. The main objective of this case series and systematic review was to share our multi-institutional experience and to assess the feasibility and safety of endovascular treatment during pregnancy, as well as the corresponding maternal and fetal outcomes based on currently available evidence.MethodsWe report a case series of 12 pregnant women presenting with hemorrhagic stroke from ruptured cerebrovascular arteriovenous malformations or aneurysms who underwent endovascular treatment prior to delivery. A systematic literature review of pregnant patients with endovascular treated cerebrovascular malformations, published between 1995 and 2022, was performed. Clinical patient information, detailed treatment strategies, maternal and fetal outcomes as well as information on the delivery were collected and assessed.ResultsIn most patients the course was uneventful and an excellent outcome without significant neurological deficits (mRS ≤ 1) was achieved. Furthermore, the maternal outcome was not worse compared to the general population who underwent endovascular treatment of ruptured vascular brain lesions. Also, in most cases a healthy fetus was born.ConclusionEndovascular treatment of ruptured cerebrovascular malformations during pregnancy is safe and feasible regarding both aspects, the maternal and fetal outcomes. Still, a stronger knowledge base is needed to correctly approach future cases of intracranial hemorrhage in the pregnant population.

Observational Study of Patients Hospitalized With Neurologic Events After SARS-CoV-2 Vaccination, December 2020-June 2021.

The global spread of the COVID-19 pandemic accelerated the vaccine development time line, regulatory approval, and widespread implementation in the population underscoring the importance of postauthorization/postlicensure vaccine safety surveillance. To monitor for vaccine-related adverse events, we prospectively identified patients hospitalized for prespecified neurologic conditions who received mRNA or adenovirus COVID-19 vaccines and assessed cases for potential risk factors and alternative etiologies of the adverse event. We identified prespecified neurologic conditions in hospitalized individuals within 6 weeks of receipt of a dose of any COVID-19 vaccination between December 11, 2020, and June 22, 2021 (Columbia University Irving Medical Center/New York Presbyterian Hospital, New York City, New York). Clinical data from electronic medical records in these vaccinated patients were reviewed for assessment of contributing risk factors and etiologies for these neurologic conditions by use of a published algorithm. Among 3,830 individuals screened for COVID-19 vaccination status and neurologic conditions, 138 (3.6%) cases were included in this study (126 after mRNA and 6 after Janssen vaccines). The 4 most prevalent neurologic syndromes included ischemic stroke (52, 37.7%), encephalopathy (45, 32.6%), seizure (22, 15.9%), and intracranial hemorrhage (ICH) (13, 9.4%). All 138 cases (100%) had 1 or more risk factors and/or evidence for established causes. Metabolic derangement was the most common etiology for seizures (24, 53.3%) and encephalopathy (5, 22.7%) while hypertension was the most significant risk factor in ischemic stroke (45, 86.5%) and ICH cases (4, 30.8%). All cases in this study were determined to have at least 1 risk factor and/or known etiology accounting for their neurologic syndromes. Our comprehensive clinical review of these cases supports the safety of mRNA COVID-19 vaccines.

Complications of Cerebral Spinal Fluid Drainage in Thoracoabdominal Aortic Procedures

Spinal cord ischemia is a devastating complication of thoracoabdominal aortic procedures. Prophylactic cerebral spinal fluid (CSF) drainage has been shown to reduce the rates of postoperative neurologic deficit. Recent articles have cited a high rate of complications secondary to CSF drain placement, ranging from 7.6% to 9%, with the most worrisome being intracranial hemorrhage (ICH) and have concluded that the risks of drain use may outweigh the benefits. A single-center retrospective cohort study was conducted to investigate our rates of CSF drain-related complications. Patients were included if they underwent a thoracic or thoracoabdominal aortic procedure and had a CSF drain placed in the perioperative period. Patients with pre-existing neurologic deficits or who presented with aortic rupture were excluded. Major complications were defined as permanent paraplegia, ICH, epidural hematoma or abscess, meningitis, or catheter retention requiring a reoperation. Minor complications that were tracked included transient parasthesia, CSF leak, post-dural puncture headache, asymptomatic blood in the CSF, drain failure, and catheter retention not requiring a reoperation. There were 129 patients in the 15-year period who received a CSF drain for spinal cord protection (Table I). Of these, 122 patients (94.6%) had prophylactic CSF drains, 4 had drains placed postoperatively, and 3 patients had both preoperative and postoperative drains. The most common indication for surgery was thoracoabdominal aortic aneurysm (n = 57; 44.2%) (Table I), and the majority of cases were performed endovascularly (n = 72; 55.8%). There were five occurrences of permanent paraplegia for a rate of 3.9% (Table II). There were no cases of ICH nor other major complications due to spinal drain placement. The rate of minor complications was 26.4%, with the most common being transient parasthesia (n = 13; 10.1%) (Table II). No association was found between complication rates and either indication for procedure or type of operation. The rate of minor complications in this cohort is comparable to prior literature, but our major complication rates due to spinal drain insertion were found to be lower, with no incidence of ICH. The rate of permanent paraplegia in the cohort was 3.9%, which is similar to previous studies. The lack of spinal drain placement complications may be due to a well-defined spinal drain protocol at our center and the small group of dedicated anesthesiologists working with vascular surgery. Further research is required to determine which patients benefit most from prophylactic CSF drainage.Table IPatient characteristicsFull cohort (N = 129)Basic demographicsAge, mean (SD), years65.7 (12.3)Gender, No. (%) Male73 (56.6) Female56 (43.4)BMI, mean (SD)29.1 (6.4)Comorbidities, No. (%) Hypertension101 (78.3) Dyslipidemia60 (46.5) Coronary artery disease30 (23.3) COPD29 (22.4) Chronic kidney disease27 (20.9) Diabetes mellitus20 (15.5)Smoking history, No. (%) Current smoker44 (34.1) Prior smoker52 (40.3)Previous aortic operation, No. (%)64 (49.6)Aortic characteristics, No. (%) TAAA57 (44.2)Extent I25 (43.8)Extent II5 (8.8)Extent III16 (28.1)Extent IV11 (19.3)Mean size (SD), cm6.3 (1.2) Acute aortic syndrome, No. (%)40 (31.0)Acute dissection18 (45.0)PAU13 (32.5)IMH9 (22.5)Chronic dissection, No. (%)31 (24.0)Other indication, No. (%)1BMI, Body mass index; COPD, chronic obstructive pulmonary disease; IMH, intramural hematoma; PAU, penetrating aortic ulcer; SD, standard deviation; TAAA, thoracoabdominal aortic aneurysm. Open table in a new tab Table IIStudy outcomesFull cohort (n = 129)TAAA (n = 57)AAS (n = 40)Chronic dissection (n = 31)Major complications, No. (%)5 (3.9)0 (0)3 (7.5)2 (6.5) Permanent paraplegia5 (3.9)0 (0)3 (7.5)2 (6.5) Epidural hematoma0 (0)0 (0)0 (0)0 (0) Epidural abscess0 (0)0 (0)0 (0)0 (0) ICH0 (0)0 (0)0 (0)0 (0) Meningitis0 (0)0 (0)0 (0)0 (0) Catheter reoperation0 (0)0 (0)0 (0)0 (0)Minor complications, No. (%)34 (26.4)14 (24.6)12 (30.0)8 (25.8) Temporary parasthesia13 (10.1)7 (12.3)4 (10.0)2 (6.5) Asymptomatic bleed8 (6.2)1 (1.8)3 (7.5)4 (12.9) PDPH6 (4.7)3 (5.3)2 (5.0)1 (3.2) Drain failure5 (3.9)2 (3.5)2 (5.0)1 (3.2) CSF leak2 (1.6)1 (1.8)1 (2.5)0 (0) Catheter nonoperative0 (0)0 (0)0 (0)0 (0)AAS, Acute aortic syndrome; CSF, cerebrospinal fluid; ICH, intracranial hemorrhage; PDPH, post-dural puncture headache; TAAA, thoracoabdominal aortic aneurysm. Open table in a new tab

Conventional MR Imaging in Trauma Management in Adults.

MR imaging has been shown to have higher sensitivity than computed tomography (CT) for traumatic intracranial soft tissue injuries as well as most cases of intracranial hemorrhage, thus making it a significant adjunct to CT in the management of traumatic brain injury, mostly in the subacute to chronic phase, but may also be of use in the acute phase, when there are persistent neurologic symptoms unexplained by prior imaging.

Medical malpractice and intracranial hemorrhages in the U.S.: An analysis of 121 cases over 35 years

IntroductionThough all physicians are at risk for medical malpractice litigation, those in surgical specialties, particularly neurosurgeons, are at increased risk. Because intracranial hemorrhages are a life-threatening and commonly misdiagnosed condition, the aim of this study is to identify and increase awareness of factors associated with litigation in cases of intracranial hemorrhages. MethodsThe online legal database Westlaw was utilized to query public litigation cases related to the management of intracranial hemorrhages between 1985 and 2020. Various search terms were used to identify cases, and the following variables were extracted: plaintiff demographics, defendant specialty, trial year, court type, location, reason for litigation, plaintiff medical complaints, trial outcomes, and payouts for both verdicts and settlements. Comparative analysis was performed between cases decided in favor of the plaintiff and in favor of the defendant. ResultsA total of 121 cases met inclusion criteria. The most common type of hemorrhage was subarachnoid (65.3%), and the most common cause of hemorrhage was cerebral aneurysm/vascular malformation (37.2%). Most cases were brought against a hospital or healthcare system (60.3%), followed by emergency medicine physicians (33.1%), family medicine physicians (10.7%), and neurosurgeons (6.6%). Failure to diagnose was the most common reason for litigation (84.3%). Cases most frequently resulted in verdicts favoring the defense (48.8%), followed by settlements (35.5%). Plaintiffs were found to be significantly younger in cases ruled in favor of the plaintiff than in cases ruled in favor of the defense (p = 0.014). Cases ruled in favor of the plaintiff were also significantly more likely to involve a neurologist (p = 0.029). ConclusionsMost cases of intracranial hemorrhage resulting in malpractice litigation were classified as subarachnoid hemorrhages and caused by aneurysm/vascular malformation. Most cases were brought against hospital systems, and failure to diagnose was the most common reason for litigation. Cases resulting in verdicts in favor of the plaintiff were significantly more likely to involve younger plaintiffs and neurologists.

Twenty years' experience of immune thrombocytopenia and intracranial haemorrhage in the Nordic countries-A NOPHO study.

Twenty years' experience of immune thrombocytopenia and intracranial haemorrhage in the Nordic countries-A NOPHO study.

Intraventricular hemorrhage following thrombolytic therapy with streptokinase: A case series

Streptokinase is a widely used medication for the purpose of thrombolysis in case of acute coronary artery thrombosis despite being high risks of bleeding complications. The main reason behind it may be that it is one of the immediate lifesaving measure in cases of myocardial infarction. Keeping in mind, the post-thrombolytic measures in all such patients are need of an hour prevent such complicators. We hereby report three such cases of intracranial hemorrhage in patients undergoing thrombolytic therapy.

The results of the use of thrombolytic therapy in acute myocardial infarction at the prehospital and hospital stage

The aim of the study was study of the use of thrombolytic therapy (TLT) in patients with myocardial infarction (MI) with ST segment elevationat the prehospital and hospital stage. The study included 181 patients with ST elevation MI. Depending on the thrombolytic used, the patients were divided into two groups: patients treated with streptokinase were included in the 1st group (n=78); the 2nd group (n=51) included patients with MI with ST segment elevation, who did not undergo TLT due to late seeking medical help or the presence of contraindications. TLT was performed both at the prehospital stage and in a hospital setting. We analyzed the time before TLT, the clinical course of MI, mortality, and complications. The time from the onset of clinical symptoms was 2.70.22 hours. The high efficiency of these thrombolytics was established in 66% of patients. Mortality during TLT was 6.4–7.7%, without TLT - 24%. Carrying out TLT in the first 3 hours from the onset of a heart attack was accompanied by a decrease in mortality to 3.4%. Not a single case of intracranial hemorrhage and allergic reactions was recorded.± TLT with streptokinase in patients with MI with ST segment elevation on the ECG in real clinical practice was characterized by high efficiency, a decrease in hospital mortality, complications, and a low percentage of side effects.

Fetal intracranial hemorrhage: prenatal sonographic diagnosis criteria and postnatal outcomes

The aim of this study was to improve knowledge of prenatally diagnosed fetal intracranial hemorrhage (ICH), defining the ultrasound (US) examination results, the contribution of fetal magnetic resonance imagination (MRI) to the diagnosis, and the pregnancy outcomes, from a series of fetal ICH cases. This retrospective, observational study included eleven fetuses diagnosed with ICH from April 2016 to August 2020. The data regarding the medical records, prenatal US and MRI findings, treatment, and prognosis of fetal ICH cases were collected from the hospital database and analyzed. Fetal ICHs were grade 3 in six cases, and grade 4 in the remaining five cases. The mean gestational age at diagnosis was 30.2 weeks. Nine (81.8%) of the cases were diagnosed in the third trimester and two (18.2%) in the second trimester. Fetal cranial MRI was performed in 7/11 (63.6%) following ultrasonographic diagnosis. MRI confirmed fetal ICH diagnosis and previous US findings regarding location and grade in all cases. Five patients (45.5%) diagnosed with grade 3 (n=1) and grade 4 (n=4) ICH underwent pregnancy termination. Of the remaining six cases, one (9.1%) diagnosed with grade 3 fetal ICH resulted in an intrauterine fetal demise. Four cases classified as grade 3 fetal ICH and one case with grade 4 fetal ICH were born alive at term. The clinical manifestations of fetal ICH are diverse and have a wide spectrum of severity and prognostic implications. Fetal ICH cases were mainly detected in the third trimester, with a minority detected in the second trimester. These cases can be safely diagnosed and graded by US examination, but the underlying etiology frequently cannot be determined. Fetal cranial MRI may aid in diagnosis confirmation if this is unclear from US in order to provide appropriate counseling to the parents.

Improved performance and robustness of multi-task representation learning with consistency loss between pretexts for intracranial hemorrhage identification in head CT.

With the recent development of deep learning, the classification and segmentation tasks of computer-aided diagnosis (CAD) using non-contrast head computed tomography (NCCT) for intracranial hemorrhage (ICH) has become popular in emergency medical care. However, a few challenges remain, such as the difficulty of training due to the heterogeneity of ICH, the requirement for high performance in both sensitivity and specificity, patient-level predictions demanding excessive costs, and vulnerability to real-world external data. In this study, we proposed a supervised multi-task aiding representation transfer learning network (SMART-Net) for ICH to overcome these challenges. The proposed framework consists of upstream and downstream components. In the upstream, a weight-shared encoder of the model is trained as a robust feature extractor that captures global features by performing slice-level multi-pretext tasks (classification, segmentation, and reconstruction). Adding a consistency loss to regularize discrepancies between classification and segmentation heads has significantly improved representation and transferability. In the downstream, the transfer learning was conducted with a pre-trained encoder and 3D operator (classifier or segmenter) for volume-level tasks. Excessive ablation studies were conducted and the SMART-Net was developed with optimal multi-pretext task combinations and a 3D operator. Experimental results based on four test sets (one internal and two external test sets that reflect a natural incidence of ICH, and one public test set with a relatively small amount of ICH cases) indicate that SMART-Net has better robustness and performance in terms of volume-level ICH classification and segmentation over previous methods. All code is available at https://github.com/babbu3682/SMART-Net.

Analysis of non-retinopathy of prematurity (ROP)-related fundus hemorrhage in preterm infants in China.

ObjectiveTo determine the incidence of fundus hemorrhage (FH) not associated with retinopathy of prematurity (ROP) during ocular screening and report their clinical features, risk factors, therapies, and prognosis in a large population of premature newborns.MethodsThe medical records of all preterm newborns admitted to the Department of Ophthalmology, Peking University People's Hospital, Beijing, from January 1, 2016 through August 31, 2021 were retrospectively reviewed. Fundus examinations were carried out by experienced retinal experts. Examination under anesthesia was carried out in newborns with abnormal fundus including vitreous hemorrhage (VH) or retinal hemorrhage (RH) >2 disks' diameter by a Retcam 2 system. A lens-preserving vitrectomy was performed in infants requiring a vitrectomy. A comprehensive medical history was also recorded and analyzed.ResultsDuring the 5-year period, a total of 7,260 preterm babies were screened. There were 82 (1.13%) newborns and 104 (0.72) eyes with FH, including VH or RH.Twelve (14.63%) newborns (16 eyes, 15.38%) had VH; 56 (68.29%) (74 eyes, 71.15%) had flame-shaped, superficial hemorrhages; six (7.31%) (6 eyes, 5.77%) had small, round, deep hemorrhages (<2 disk diameters in size); and eight (9.76%) (8 eyes, 7.69%) had large, round hemorrhages (>2 disk diameters). In all, there were 10 (12.20%) cases of intracranial hemorrhage. The mode of delivery was not found to be a significant factor in the occurrence of birth-related retinal hemorrhage (P = 0.22).Six newborns (eyes) with large, round retinal hemorrhage at the posterior pole while the macular was not impacted and 11 cases (15 eyes) with vitreous hemorrhage were required to receive close follow-up with average follow-up time of 105 days. A lens-sparing vitreous surgery was conducted in three patients without any complications.ConclusionPreterm newborns with FH that are not caused by ROP are more likely to have superficial, peripheral hemorrhages. Vaginal delivery compression and forceps may be associated with hemorrhage. A lens-preserving vitrectomy is required and considered safe for dense FH involving the refractive media.