-
Cureus Apr 2024Background Pulse oximetry screening (POS) is acknowledged globally as a noninvasive method to detect critical congenital heart diseases (CCHDs) and respiratory...
Background Pulse oximetry screening (POS) is acknowledged globally as a noninvasive method to detect critical congenital heart diseases (CCHDs) and respiratory illnesses. However, its value for early diagnosis and treatment remains unrecognized in many hospitals with limited resources around the world. This study aimed to evaluate POS's application in CCHDs, persistent pulmonary hypertension (PPHN), and respiratory distress syndrome (RDS) for early diagnosis and its influence on clinical procedures in rural areas. Methods This prospective observational study included all eligible newborn infants in the regional neonatal unit of a community healthcare center. Their peripheral oxygen saturation was assessed at <24 hours and >24 hours after birth, in the right upper limb and either lower limb. An oxygen saturation of <95% or >3% difference between pre-ductal and post-ductal circulations was considered abnormal. All neonates with abnormal oxygen saturations at >24 hours after birth were subjected to another POS test within two hours of the last test. If the oxygen saturation was still abnormal, it was considered a positive POS test. The POS results were classified as oxygen saturation abnormal (<90%), abnormal (90-94%), and normal (≥95%). All neonates with a positive POS test were referred for echocardiography. Results Overall, 440 infants had documented POS results. A total of 65 (14.77%) infants had a positive POS test result, out of which 39 (8.86%) cases were diagnosed on further evaluation. Four neonates had CCHD (positive predictive value (PPV) = 6.15%), 26 had RDS (PPV = 40%), and nine had PPHN (PPV = 13.85%). Without any further delay, the doctor directed them all to a more advanced facility. Conclusion Our research showed that, in large-scale clinical settings, the addition of pulse oximetry to routine cardiac auscultation could be a reliable and feasible method to screen newborns for CCHD, PPHN, and RDS early on. Our research underscores the importance of implementing routine POS to detect CCHD, RDS, and PPHN in clinical practice.
PubMed: 38756257
DOI: 10.7759/cureus.58398 -
NPJ Digital Medicine May 2024Healthcare datasets are becoming larger and more complex, necessitating the development of accurate and generalizable AI models for medical applications. Unstructured...
Healthcare datasets are becoming larger and more complex, necessitating the development of accurate and generalizable AI models for medical applications. Unstructured datasets, including medical imaging, electrocardiograms, and natural language data, are gaining attention with advancements in deep convolutional neural networks and large language models. However, estimating the generalizability of these models to new healthcare settings without extensive validation on external data remains challenging. In experiments across 13 datasets including X-rays, CTs, ECGs, clinical discharge summaries, and lung auscultation data, our results demonstrate that model performance is frequently overestimated by up to 20% on average due to shortcut learning of hidden data acquisition biases (DAB). Shortcut learning refers to a phenomenon in which an AI model learns to solve a task based on spurious correlations present in the data as opposed to features directly related to the task itself. We propose an open source, bias-corrected external accuracy estimate, P, that better estimates external accuracy to within 4% on average by measuring and calibrating for DAB-induced shortcut learning.
PubMed: 38744921
DOI: 10.1038/s41746-024-01118-4 -
Alternative Therapies in Health and... May 2024Postoperative pneumonia in critically ill patients is becoming an important cause for adverse clinical outcomes. It is very important to predict postoperative pneumonia....
OBJECTIVE
Postoperative pneumonia in critically ill patients is becoming an important cause for adverse clinical outcomes. It is very important to predict postoperative pneumonia. Surgical Intensive Care Unit(SICU), is an intensive care unit that deals with post-surgical patients, and is usually staffed by a team of surgeons, critical care specialists, and nurses to provide close monitoring and care. The purpose of this study is to investigate the risk factors of postoperative pneumonia in patients in SICU after surgery, establish a risk prediction model, and help surgeons and SICU doctors to early identify patients with high-risk postoperative pneumonia.
METHODS
To explore risk factors for postoperative pneumonia, Patients in the SICU from January 1, 2019, to December 31, 2019, were collected and retrospectively analyzed. The data were randomly divided into a derivation set (n=533) and a validation set (n=277). Patients were divided into postoperative pneumonia (PP) group and non-postoperative pneumonia (NPP) group. t test and Chi-square test were used to compare the differences between the PP and NPP groups before and after surgery. The risk factors of postoperative pneumonia in SICU patients were identified using univariate and multivariate logistic regression. A derivation set was used to build the model, and a validation set was used for model evaluation. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) were used to evaluate the model performance. The model was validated by AUC using a validation set.
RESULTS
With this model, a total of 8 independent risk factors were identified to be associated with postoperative pneumonia in SICU patients after surgery. Patients with the 8 risk factors were assigned the following scores: recorded aspiration: 8, preoperative disturbance of consciousness: 4, thoracic and abdominal surgery: 3, contaminated wound: 10, abnormal choking cough on SICU admission: 9, abnormal pulmonary auscultation on SICU admission: 5, postoperative sedation, 4 points, and postoperative analgesia >1 day: 3. Eight risk factors were significantly correlated with postoperative pneumonia. Based on the scoring standard above, a risk factor table was created using the 8 predictors with a total score of 46. The AUC was 0.933 and 0.908 in derivation set and validation set. A cumulative score > 12 indicates high risk of postoperative pneumonia.
CONCLUSIONS
This study identified 8 risk factors that are significantly associated with postoperative pneumonia in SICU patients after surgery and provides operable clinical tools for early prevention and intervention of postoperative pneumonia. The implementation of this model has significant potential to enhance patient outcomes in the SICU by enabling early identification and stratification of patients at elevated risk of developing postoperative pneumonia. This model allows for the timely initiation of targeted preventative and therapeutic interventions, potentially reducing the incidence of pneumonia, shortening hospital stays, and improving overall patient survival rates. Furthermore, the use of a cumulative scoring system, simplifies clinical decision-making, making it accessible and actionable for surgeons and SICU staff.
PubMed: 38743891
DOI: No ID Found -
The American Journal of Medicine May 2024Few studies have assessed the ability of internal medicine residents to perform a cardiovascular physical examination using real patients.
BACKGROUND
Few studies have assessed the ability of internal medicine residents to perform a cardiovascular physical examination using real patients.
METHODS
First year internal medicine interns from 2 large academic medical centers in Maryland examined the same patient with aortic insufficiency as part of the Assessment of Physical Examination and Communication Skills (APECS). Interns were assessed on 5 clinical domains: physical exam technique, identifying physical signs, generating a differential diagnosis, clinical judgment, and maintaining patient welfare. Spearman's correlation test was used to describe associations between clinical domains. Preceptor comments were examined to identify common errors in physical exam technique and identifying physical signs.
RESULTS
One hundred nine interns examined the same patient with aortic insufficiency across 14 APECS sessions. Only 58 interns (53.2%) correctly identified the presence of a diastolic murmur, and only 52 interns (47.7%) included aortic insufficiency on their differential diagnosis. There was a significant and positive correlation between physical exam technique and identification of the correct physical findings (r = 0.42, P < .001). Both technique (r = 0.34, P = .003) and identifying findings (r = 0.42, P < .001) were significantly associated with generating an appropriate differential diagnosis. Common errors in technique included auscultating over the gown, timing the cardiac cycle with the radial pulse, and failing to palpate for the apical impulse.
CONCLUSIONS
Internal medicine interns had variable skills in performing and interpreting the cardiovascular physical exam. Improving cardiovascular exam skills would likely lead to increased identification of relevant cardiovascular findings, inform clinical decision making and improve overall patient care.
PubMed: 38740321
DOI: 10.1016/j.amjmed.2024.04.039 -
Cureus Apr 2024Primary cardiac tumors (PCTs) are less frequent and carry an incidence of 1.38 per 100,000 population per year. Myxofibrosarcomas are reported as one of the rarest forms...
Primary cardiac tumors (PCTs) are less frequent and carry an incidence of 1.38 per 100,000 population per year. Myxofibrosarcomas are reported as one of the rarest forms of cardiac sarcomas, mostly with mesenchymal origin and located in the left atrium. Current research indicates an increase in median survival from 14 months to 36 months following complete resection and chemoradiotherapy. A 55-year-old Caucasian woman was admitted with brief self-resolving episodes of aphasia following migraine headaches for the past few months with associated exertional dyspnea and episodes of hypotension. Examination revealed a right-sided facial droop with cardiac murmur on auscultation. MRI brain was recommended which revealed a non-hemorrhagic infarct and multiple watershed infarcts. A transesophageal echocardiography revealed a large mass of around 5 cm in size located at the posterior wall of the left atrium causing mitral stenosis. The patient was initially managed conservatively and referred to cardiothoracic surgery and underwent a complete surgical resection. The histopathological report indicated the presence of primary cardiac sarcoma, and a postoperative positron emission therapy (PET) scan revealed no other foci of cancer further strengthening evidence of a primary cardiac pathology. This case represents a rare cardiac pathology presenting with non-cardiac symptoms.
PubMed: 38738092
DOI: 10.7759/cureus.58000 -
Animals : An Open Access Journal From... Apr 2024Auscultation of heart sounds is an important veterinary skill requiring an understanding of anatomy, physiology, pathophysiology and pattern recognition. This...
Auscultation of heart sounds is an important veterinary skill requiring an understanding of anatomy, physiology, pathophysiology and pattern recognition. This cross-sectional study was developed to evaluate a targeted, audio-visual training resource for veterinary students to improve their understanding and auscultation of common heart conditions in horses. Fourth- and fifth-year 2021 and 2022 Bachelor of Veterinary Science students at the University of Queensland (UQ) were provided the learning resource and surveyed via online pre- and post-intervention surveys. Results were quantitatively analyzed using descriptive statistics and Mann-Whitney U tests. Open-ended survey questions were qualitatively analyzed by thematic analysis and Leximancer™ Version 4 program software analysis. Over the two-year period, 231 fourth-year and 222 fifth-year veterinary students had access to the resource; 89 completed the pre-intervention survey and 57 completed the post-intervention survey. Quantitative results showed the resource helped students prepare for practicals and their perception of competency and confidence when auscultating equine cardiac sounds improved ( < 0.05). Compared to fifth-year students, fourth-year students felt less competent at identifying murmurs and arrythmias prior to accessing the learning resource ( < 0.05). Fourth-year and fifth-year students' familiarity with detection of murmurs improved after completing the learning resource ( < 0.001). Qualitative analysis demonstrated a limited number of opportunities to practice equine cardiac auscultation throughout the veterinary degree, especially during the COVID-19 pandemic, and that integrated audio-visual resources are an effective means of teaching auscultation.
PubMed: 38731348
DOI: 10.3390/ani14091341 -
Identifying pediatric heart murmurs and distinguishing innocent from pathologic using deep learning.Artificial Intelligence in Medicine Jul 2024To develop a deep learning algorithm to perform multi-class classification of normal pediatric heart sounds, innocent murmurs, and pathologic murmurs.
OBJECTIVE
To develop a deep learning algorithm to perform multi-class classification of normal pediatric heart sounds, innocent murmurs, and pathologic murmurs.
METHODS
We prospectively enrolled children under age 18 being evaluated by the Division of Pediatric Cardiology. Parents provided consent for a deidentified recording of their child's heart sounds with a digital stethoscope. Innocent murmurs were validated by a pediatric cardiologist and pathologic murmurs were validated by echocardiogram. To augment our collection of normal heart sounds, we utilized a public database of pediatric heart sound recordings (Oliveira, 2022). We propose two novel approaches for this audio classification task. We train a vision transformer on either Markov transition field or Gramian angular field image representations of the frequency spectrum. We benchmark our results against a ResNet-50 CNN trained on spectrogram images.
RESULTS
Our final dataset consisted of 366 normal heart sounds, 175 innocent murmurs, and 216 pathologic murmurs. Innocent murmurs collected include Still's murmur, venous hum, and flow murmurs. Pathologic murmurs included ventricular septal defect, tetralogy of Fallot, aortic regurgitation, aortic stenosis, pulmonary stenosis, mitral regurgitation and stenosis, and tricuspid regurgitation. We find that the Vision Transformer consistently outperforms the ResNet-50 on all three image representations, and that the Gramian angular field is the superior image representation for pediatric heart sounds. We calculated a one-vs-rest multi-class ROC curve for each of the three classes. Our best model achieves an area under the curve (AUC) value of 0.92 ± 0.05, 0.83 ± 0.04, and 0.88 ± 0.04 for identifying normal heart sounds, innocent murmurs, and pathologic murmurs, respectively.
CONCLUSION
We present two novel methods for pediatric heart sound classification, which outperforms the current standard of using a convolutional neural network trained on spectrogram images. To our knowledge, we are the first to demonstrate multi-class classification of pediatric murmurs. Multiclass output affords a more explainable and interpretable model, which can facilitate further model improvement in the downstream model development cycle and enhance clinician trust and therefore adoption.
Topics: Humans; Deep Learning; Heart Murmurs; Child; Child, Preschool; Infant; Adolescent; Prospective Studies; Heart Sounds; Female; Male; Algorithms; Diagnosis, Differential; Heart Auscultation
PubMed: 38723434
DOI: 10.1016/j.artmed.2024.102867 -
Cureus Apr 2024Intubation in emergency settings demands rapid confirmation of endotracheal tube (ETT) placement for establishing a definitive airway. Current methods, including...
Intubation in emergency settings demands rapid confirmation of endotracheal tube (ETT) placement for establishing a definitive airway. Current methods, including capnography and auscultation, have limitations. This study introduces a cost-effective and easily accessible training model for confirming ETT placement using ultrasound, aiming to improve training and patient outcomes. We developed a gelatin and psyllium-based model that simulates adult ETT intubation, offering an alternative to costly cadaveric models. The model's construction is described, with materials costing approximately $7.34 per unit. Preliminary results show promise in simulating tracheal and esophageal intubation scenarios. This novel model provides an ethical and economical solution for training healthcare professionals in the ultrasound confirmation of ETT placement, paving the way for further validation and adoption in medical education.
PubMed: 38721221
DOI: 10.7759/cureus.57830 -
Acute Coronary Syndrome at Altitude: Diagnostic Dilemma on Aconcagua Using Point-of-Care Ultrasound.Wilderness & Environmental Medicine May 2024At the Plaza de Mulas medical tent, located at 4300 m (14,100 ft) along the Normal Route to the 6960 m (22,837 ft) summit of Aconcagua in Argentina, a Korean male in...
At the Plaza de Mulas medical tent, located at 4300 m (14,100 ft) along the Normal Route to the 6960 m (22,837 ft) summit of Aconcagua in Argentina, a Korean male in his 50s with no known medical conditions presented with lightheadedness and shortness of breath. He had taken sildenafil and acetazolamide that morning without improvement. Vital signs on arrival were notable for oxygen saturations in the high 60s with basilar crackles on lung auscultation, concerning for high altitude pulmonary edema. The patient was started on oxygen via nasal cannula and given dexamethasone. History was limited secondary to language barriers, but on review of systems the patient noted mild chest pressure. Bedside cardiac echocardiogram was performed, which revealed a septal wall motion abnormality. The patient was therefore given aspirin and clopidogrel and was flown to a lower trailhead, where he was met by local Emergency Medical Services. A 12-lead electrocardiogram revealed an anterior ST-elevation myocardial infarction, and the patient was taken emergently to the catheterization lab in Mendoza and underwent stent placement with a full recovery.
PubMed: 38720618
DOI: 10.1177/10806032241249128 -
JASA Express Letters May 2024Machine learning enabled auscultating diagnosis can provide promising solutions especially for prescreening purposes. The bottleneck for its potential success is that...
Machine learning enabled auscultating diagnosis can provide promising solutions especially for prescreening purposes. The bottleneck for its potential success is that high-quality datasets for training are still scarce. An open auscultation dataset that consists of samples and annotations from patients and healthy individuals is established in this work for the respiratory diagnosis studies with machine learning, which is of both scientific importance and practical potential. A machine learning approach is examined to showcase the use of this new dataset for lung sound classifications with different diseases. The open dataset is available to the public online.
Topics: Humans; Machine Learning; Auscultation; Respiratory Sounds
PubMed: 38717466
DOI: 10.1121/10.0025851