Using Deep-Learning Algorithms to Simultaneously Identify Right and Left Ventricular Dysfunction From the Electrocardiogram

ObjectivesThis study sought to develop DL models capable of comprehensively quantifying left and right ventricular dysfunction from ECG data in a large, diverse population.BackgroundRapid evaluation of left and right ventricular function using deep learning (DL) on electrocardiograms (ECGs) can assist diagnostic workflow. However, DL tools to estimate right ventricular (RV) function do not exist, whereas those to estimate left ventricular (LV) function are restricted to quantification of very low LV function only.MethodsA multicenter study was conducted with data from 5 New York City hospitals: 4 for internal testing and 1 serving as external validation. We created novel DL models to classify left ventricular ejection fraction (LVEF) into categories derived from the latest universal definition of heart failure, estimate LVEF through regression, and predict a composite outcome of either RV systolic dysfunction or RV dilation.ResultsWe obtained echocardiogram LVEF estimates for 147,636 patients paired to 715,890 ECGs. We used natural language processing (NLP) to extract RV size and systolic function information from 404,502 echocardiogram reports paired to 761,510 ECGs for 148,227 patients. For LVEF classification in internal testing, area under curve (AUC) at detection of LVEF ≤40%, 40% < LVEF ≤50%, and LVEF >50% was 0.94 (95% CI: 0.94-0.94), 0.82 (95% CI: 0.81-0.83), and 0.89 (95% CI: 0.89-0.89), respectively. For external validation, these results were 0.94 (95% CI: 0.94-0.95), 0.73 (95% CI: 0.72-0.74), and 0.87 (95% CI: 0.87-0.88). For regression, the mean absolute error was 5.84% (95% CI: 5.82%-5.85%) for internal testing and 6.14% (95% CI: 6.13%-6.16%) in external validation. For prediction of the composite RV outcome, AUC was 0.84 (95% CI: 0.84-0.84) in both internal testing and external validation.ConclusionsDL on ECG data can be used to create inexpensive screening, diagnostic, and predictive tools for both LV and RV dysfunction. Such tools may bridge the applicability of ECGs and echocardiography and enable prioritization of patients for further interventions for either sided failure progressing to biventricular disease.     

Who Remembers Whether They Had Asthma as Children?

Purpose: To assess misclassification in adults reporting childhood asthma. Methods: The Tasmanian Asthma Survey commenced in 1968 when 8,583 7-year-old children attending school in Tasmania (participants) were studied. In 1991–1993, a stratified sample of 1,494 participants was studied. Their recall of childhood asthma was compared to their parents' prospectively gathered report. Results: Where participants had childhood asthma, those with current asthma, severe eczema, or hay fever were less likely to misclassify while females were more likely to misclassify. Where participants had no childhood asthma, misclassification was associated with current asthma, hay fever, or allergies. Conclusions: Retrospective self-assessment of childhood asthma is unreliable.     

Effect of hospital noise on patients' ability to hear,understand, and recall speech

Speech intelligibility and recall were examined in normally hearing and hearing‐impaired hospitalized patients. Fifty‐two participants completed testing in a sound‐attenuated booth. While listening to a recorded male speaker talking at conversational level, participants were asked to identify and remember the last (key) word in each of a series of five sentences presented in hospital noise with or without voices at three decibel levels (59, 64, and 69 dBA). Noise level and sentence context had the largest impact on key word identification (p < .001). Noise level had the largest impact on key word recall (p < .001). Type of hospital noise and hearing loss also significantly influenced performance on both measures. These findings have implications for healthcare providers communicating with hospitalized patients. © 2013 Wiley Periodicals, Inc. Res Nurs Health 36:228–241, 2013     

Dynamic training of the lumbar musculature to prevent recurrence of acute low back pain: a randomized controlled trial using a daily pain recall for 1 year

Purpose: The purpose of this experiment was to quantify lumbar muscle endurance training for individuals with a recent episode of acute low back pain (LBP) (≥ moderate pain for ≥ 2 days) and to observe whether the training would reduce the rate and severity of recurrent LBP episodes. Method: Twenty-six participants who were pain-free at the time of the study were randomly divided into a high intensity back endurance (HIBE)-trained or a low intensity abdominal (LOAB)-trained (control) group. The HIBE-trained group performed preloaded maximum isokinetic exertions of the back extensors (five sets of 10 repetitions, 3 days a week for 4 weeks, totaling 12 sessions). The LOAB-trained group performed low intensity isometric contractions on their abdominals that had minimal effect on their back musculature. The two groups reported daily pain logs on a weekly basis on an interactive voice response telephone system for 1 year. Results: The HIBE-trained group experienced more pain days (p = 0.038) in the minor and moderate categories and more episodes of acute LBP than the LOAB-trained group. However, there was a trend of less pain in the severe, intense and excruciating categories in the HIBE-trained group. Conclusions: The results of this experiment did not provide evidence that short-term intense training of the low back musculature provides protection against future episodes of LBP.

Implications for Rehabilitation

• Short-term high intensity back endurance (HIBE) training of the back musculature did not provide protection against future episodes of acute low back pain compared to the low intensity abdominal (LOAB) – trained group.

• HIBE training may have sensitized the subjects to report significantly more minor and moderate pain and less pain-free days compared to the LOAB-trained group.

• There was a strong trend that the HIBE-trained group experienced less pain in the severe, intense and excruciating categories compared to the LOAB-trained group.

• Future studies in rehabilitation must require daily recall of pain and quick reporting in order to capture the subtle effects training can elicit from pain reporting.

     

Improvement in accuracy of delayed recall in aged and non-aged, mature monkeys after intramuscular or transdermal administration of the CNS nicotinic receptor agonist ABT-418

 ABT-418 was evaluated for its ability to enhance accuracy on a delayed matching-to-sample (DMTS) task by aged monkeys following intramuscular administration, and in non-aged mature monkeys following transdermal application. Aged monkeys were impaired in their performance of the DMTS task such that the longest delay intervals performed at above-chance levels extended only to 20 s. In contrast, for non-aged, mature animals, delay intervals extended to 140 s. In aged monkeys, the response to ABT-418 was highly individualized with animals responding to one or more doses in the range of 2–259 nmol/kg. A systematic dose-dependent enhancement of DMTS accuracy was not observed. When the individualized ”best dose” was administered on a separate occasion, overall DMTS accuracy was increased by 12.6%. By 24 h after administration, accuracy was at control levels. In young monkeys, a significant dose-dependent enhancement of DMTS performance (an overall increase of 11.25% above baseline accuracy) was observed 5 h after application of a transdermal patch designed to maintain steady-state plasma levels of ABT-418 of 40–60 ng/ml over a 24-h period. Again there was some individual responsiveness to one of the three doses. When data included only the individualized best doses of ABT-418 for each animal, a similar enhancement of accuracy was observed for both the 5-h and 24-h test intervals. In neither the aged nor the young cohorts was enhancement of performance associated with altered response latencies or with any overt side effects of ABT-418. Thus, these data are consistent with the ability of ABT-418 to improve DMTS performance in both young and aged monkeys. In aged monkeys, this response was observed only after administration of individualized optimal doses for different monkeys. In young monkeys, a more systematic enhancement of DMTS accuracy was observed. Further, transdermal delivery of ABT-418 in non-aged monkeys demonstrated prolonged performance enhancement compared with IM injection to at least 24 h after patch administration. Received: 10 September 1996 / Final version: 7 November 1996     

Individualizing recall intervals in child dental care

Abstract Individualizing and extending recall intervals for children have been recommended in Norway to target resources efficiently. Recall intervals were changed for children aged 3–18 years in Drammen from 1991. Clinical time spent by dentists and dental hygienists, dental health status and length of recall intervals were registered from 1990 to 1993. For the child population, the mean recall interval changed from 12.5 to 13.7 months and the annual time spent per child was reduced by 14% from 1990–91 to 1992–93. Adjusted for the decline in number of new decayed teeth, the reduction in time spent was 11%. Children with intervals of 17 to 20 months had fewer new decayed teeth and their care required less personnel time than other children. For children with new decayed teeth, time spent for dental care was not associated with recall interval, while for children without new decayed teeth, longer recall intervals were associated with shorter time for dental care. The variation in number of decayed teeth and time spent for dental care was substantial at all intervals. Individualizing and extending recall intervals to some extent targeted resources at children with more dental disease. However, in the short run, inequality in dental health persisted. Limited extension of recall intervals did not interrupt the long-term trend toward better dental health in the children and substantial resources were saved in the dental services.     

Marijuana: Dose effects on pulse rate,subjective estimates of intoxication,free recall and recognition memory

The effect of marijuana on memory as measured by free recall and recognition, pulse rate and self ratings of intoxication was evaluated in 16 male volunteers. Marijuana containing 0, 5, 10 or 15 mg Δ⁹-THC was evaluated to all subjects by smoking in 4 sessions separated by a 1 week interval. Free recall was reduced in a dose related manner by the drug, but recognition memory was unaffected. A 2 sec word presentation rate produced inferior recall in comparison to a 4 sec rate, but this variable did not interact with drug condition. Intrusion errors increased following intoxication but this effect was not systematically related to dosage of Δ⁹-THC. Both pulse ratings of intoxication increased with dosage.     

Radiation recall phenomenon secondary to capecitabine: possible role of thymidine phosphorylase

Background: The first reported case of radiation (XRT) recall related to capecitabine described dermatitis in a previously radiated field in a breast cancer patient (Ortman; JCO). We previously reported the first case of recall syndrome manifesting as diffuse gastritis and duodenitis related to capecitabine with prior XRT with 5-FU in a pancreatic cancer patient (Saif; JARCET). We report here another pancreatic cancer patient with a radiation recall receiving capecitabine following capecitabine-XRT. Patients and methods: From April 2004 to June 2005, 20 patients with locally advanced pancreatic adenocarcinoma were treated with capecitabine 1,600 mg/m² daily with concomitant radiation (5040cGy) Monday–Friday (weekends off) for a total of 6 weeks, followed by capecitabine 2,000 mg/m² daily for 14 days every 3 weeks. One male patient with tumor in the neck and body of pancreas and not infiltrating the duodenum dropped hemoglobin to 7.3 g/dl at the end of the ninth week, and melena on rectal examination. Specimen of primary pancreatic ductal adenocarcinoma was obtained via EUS-guided biopsy before starting XRT on day 1 and utilized for RNA extraction. TP mRNA level was determined by real-time quantitative PCR (RT-Q-PCR). Results: Upper endoscopy revealed gastritis consistent with radiation toxicity. Colonoscopy was negative. Transfusion of three units of packed red blood cells (PRBCs) was given. The dose of capecitabine was reduced by 25%. His anemia continued to progress, a CT scan revealed 39% decrease in the tumor size (PR). Analysis of tumor specimen prior to the start of capecitabine-XRT showed TP expression of 183.16 (high). In addition to TP, DPD was 7.40, and TNF-alpha 4,114.56. Conclusion: We believe this case to be the second case of radiation recall presenting as diffuse gastritis in a patient receiving capecitabine after previous treatment with XRT. Further studies, including the role of TP are warranted into the pathogenesis of this unique phenomenon.     

Validation of short term recall of mobile phone use for the Interphone study

Aim

To validate short term recall of mobile phone use within Interphone, an international collaborative case control study of tumours of the brain, acoustic nerve, and salivary glands related to mobile telephone use.

Methods

Mobile phone use of 672 volunteers in 11 countries was recorded by operators or through the use of software modified phones, and compared to use recalled six months later using the Interphone study questionnaire. Agreement between recalled and actual phone use was analysed using both categorical and continuous measures of number and duration of phone calls.

Results

Correlations between recalled and actual phone use were moderate to high (ranging from 0.5 to 0.8 across countries) and of the same order for number and duration of calls. The kappa statistic demonstrated fair to moderate agreement for both number and duration of calls (weighted kappa ranging from 0.20 to 0.60 across countries). On average, subjects underestimated the number of calls per month (geometric mean ratio of recalled to actual = 0.92, 95% CI 0.85 to 0.99), whereas duration of calls was overestimated (geometric mean ratio = 1.42, 95% CI 1.29 to 1.56). The ratio of recalled to actual use increased with level of use, showing underestimation in light users and overestimation in heavy users. There was substantial heterogeneity in this ratio between countries. Inter‐individual variation was also large, and increased with level of use.

Conclusions

Volunteer subjects recalled their recent phone use with moderate systematic error and substantial random error. This large random error can be expected to reduce the power of the Interphone study to detect an increase in risk of brain, acoustic nerve, and parotid gland tumours with increasing mobile phone use, if one exists.     

What is a clinician to do-believe the patient or her urinary diary?

PURPOSE: We determined if patient recall of incontinence episodes correlates with urinary diary record. MATERIALS AND METHODS: Women with 1 or more urge incontinence episode per week completed 2, 7-day diaries, the Urinary Distress Inventory and Incontinence Impact Questionnaire, and responded to 2 recall questions. RESULTS: The median number of incontinence episodes participants recalled (6.5, 5) was higher than those recorded in the diary (1.9, 1.1) at both points. Incontinence episodes in 2, 7-day diaries correlated strongly (rho = 0.921, p <0.005) while participant recall of incontinence episodes correlated weakly (rho = 0.309, p <0.059). When subjects reported being only slightly or not bothered by urge incontinence, recall and diary record correlated strongly (rho = 0.812, p = 0.014). With increasing bother (moderate or great), recall and diary were not significantly correlated (rho = 0.528, p = 0.115). CONCLUSIONS: Women with urge incontinence either overestimate or under record incontinence episode frequency in the urinary diary. This effect is more pronounced in women who are more bothered by incontinence.