The CirCor DigiScope Dataset: From Murmur Detection to Murmur Classification.

Cardiac auscultation is one of the most cost-effective techniques used to detect and identify many heart conditions. Computer-assisted decision systems based on auscultation can support physicians in their decisions. Unfortunately, the application of such systems in clinical trials is still minimal since most of them only aim to detect the presence of extra or abnormal waves in the phonocardiogram signal, i.e., only a binary ground truth variable (normal vs abnormal) is provided. This is mainly due to the lack of large publicly available datasets, where a more detailed description of such abnormal waves (e.g., cardiac murmurs) exists. To pave the way to more effective research on healthcare recommendation systems based on auscultation, our team has prepared the currently largest pediatric heart sound dataset. A total of 5282 recordings have been collected from the four main auscultation locations of 1568 patients, in the process, 215780 heart sounds have been manually annotated. Furthermore, and for the first time, each cardiac murmur has been manually annotated by an expert annotator according to its timing, shape, pitch, grading, and quality. In addition, the auscultation locations where the murmur is present were identified as well as the auscultation location where the murmur is detected more intensively. Such detailed description for a relatively large number of heart sounds may pave the way for new machine learning algorithms with a real-world application for the detection and analysis of murmur waves for diagnostic purposes.

Microcephaly in north-east Brazil: a retrospective study on neonates born between 2012 and 2015.

OBJECTIVE: To assess the number of children born with microcephaly in the State of Paraíba, north-east Brazil. METHODS: We contacted 21 maternity centres belonging to a paediatric cardiology network, with access to information regarding more than 100 000 neonates born between 1 January 2012 and 31 December 2015. For 10% of these neonates, nurses were requested to retrieve head circumference measurements data from delivery-room books. We used three separate criteria to classify whether a neonate had microcephaly: (i) the Brazilian Ministry of Health proposed criterion: term neonates (gestational age ≥ 37 weeks) with a head circumference of less than 32 cm; (ii) Fenton curves: neonates with a head circumference of less than -3 standard deviation for age and gender; or (iii) the proportionality criterion: neonates with a head circumference of less than ((height/2))+10) ± 2. FINDINGS: Between 1 and 31 December 2015, nurses obtained data for 16 208 neonates. Depending on which criterion we used, the number of neonates with microcephaly varied from 678 to 1272 (4.2-8.2%). Two per cent (316) of the neonates fulfilled all three criteria. We observed temporal fluctuations of microcephaly prevalence from late 2012. CONCLUSION: The numbers of microcephaly reported here are much higher than the 6.4 per 10 000 live births reported by the Brazilian live birth information system. The results raise questions about the notification system, the appropriateness of the diagnostic criteria and future implications for the affected children and their families. More studies are needed to understand the epidemiology and the implications for the Brazilian health system.

A telemedicine network for remote paediatric cardiology services in north-east Brazil.

PROBLEM: Providing health care for children with congenital heart diseases remains a major challenge in low- and middle-income countries. APPROACH: In October 2011, the Government of Paraíba, Brazil, established a paediatric cardiology network in partnership with the nongovernmental organization Círculo do Coração. A cardiology team supervised all network activities, using the Internet to keep in contact with remote health facilities. The network developed protocols for screening heart defects. Echocardiograms were performed by physicians under direct online supervision of a cardiologist; alternatively, a video recording of the examination was subsequently reviewed by a cardiologist. Cardiovascular surgeons came to a paediatric hospital in the state capital once a week to perform heart surgeries. LOCAL SETTING: Until 2011, the State of Paraíba had no structured programme to care for children with heart disease. This often resulted in missed or late diagnosis, with adverse health consequences for the children. RELEVANT CHANGES: From 2012 to 2014, 73,751 babies were screened for heart defects and 857 abnormalities were identified. Detection of congenital heart diseases increased from 4.09 to 11.62 per 1000 live births (P < 0.001). Over 6000 consultations and echocardiograms were supervised via the Internet. Time to diagnosis, transfers and hospital stays were greatly reduced. A total of 330 operations were carried out with 6.7% (22/330) mortality. LESSONS LEARNT: Access to an echocardiography machine with remote supervision by a cardiologist improves the detection of congenital heart disease by neonatologists; virtual outpatient clinics facilitate clinical management; the use of Internet technology with simple screening techniques allows resources to be allocated more efficiently.