Showing papers by "Derek Bell published in 2017"

Computer-Automated Malaria Diagnosis and Quantitation Using Convolutional Neural Networks

Abstract: The optical microscope remains a widely-used tool for diagnosis and quantitation of malaria. An automated system that can match the performance of well-trained technicians is motivated by a shortage of trained microscopists. We have developed a computer vision system that leverages deep learning to identify malaria parasites in micrographs of standard, field-prepared thick blood films. The prototype application diagnoses P. falciparum with sufficient accuracy to achieve competency level 1 in the World Health Organization external competency assessment, and quantitates with sufficient accuracy for use in drug resistance studies. A suite of new computer vision techniques-global white balance, adaptive nonlinear grayscale, and a novel augmentation scheme-underpin the system's state-of-the-art performance. We outline a rich, global training set; describe the algorithm in detail; argue for patient-level performance metrics for the evaluation of automated diagnosis methods; and provide results for P. falciparum.

Development and clinical performance of high throughput loop-mediated isothermal amplification for detection of malaria.

Abstract: Background Accurate and efficient detection of sub-microscopic malaria infections is crucial for enabling rapid treatment and interruption of transmission. Commercially available malaria LAMP kits have excellent diagnostic performance, though throughput is limited by the need to prepare samples individually. Here, we evaluate the clinical performance of a newly developed high throughput (HTP) sample processing system for use in conjunction with the Eiken malaria LAMP kit. Methods The HTP system utilised dried blood spots (DBS) and liquid whole blood (WB), with parallel sample processing of 94 samples per run. The system was evaluated using 699 samples of known infection status pre-determined by gold standard nested PCR. Results The sensitivity and specificity of WB-HTP-LAMP was 98.6% (95% CI, 95.7–100), and 99.7% (95% CI, 99.2–100); sensitivity of DBS-HTP-LAMP was 97.1% (95% CI, 93.1–100), and specificity 100% against PCR. At parasite densities greater or equal to 2 parasites/μL, WB and DBS HTP-LAMP showed 100% sensitivity and specificity against PCR. At densities less than 2 p/μL, WB-HTP-LAMP sensitivity was 88.9% (95% CI, 77.1–100) and specificity was 99.7% (95% CI, 99.2–100); sensitivity and specificity of DBS-HTP-LAMP was 77.8% (95% CI, 54.3–99.5) and 100% respectively. Conclusions The HTP-LAMP system is a highly sensitive diagnostic test, with the potential to allow large scale population screening in malaria elimination campaigns.

Detection of Plasmodium knowlesi, Plasmodium falciparum and Plasmodium vivax using loop-mediated isothermal amplification (LAMP) in a co-endemic area in Malaysia.

Abstract: Plasmodium knowlesi is the most common cause of malaria in Malaysia. However, microscopic diagnosis is inaccurate and rapid diagnostic tests (RDTs) are insufficiently sensitive. PCR is sensitive and specific but not feasible at a district level. Loop-mediated isothermal amplification (LAMP) shows potential with only basic requirements. A commercially available LAMP assay, the Eiken Loopamp™ MALARIA Pan Detection kit, is sensitive for Plasmodium falciparum and Plasmodium vivax, but has not previously been evaluated for P. knowlesi. This study aims to determine the sensitivity of this LAMP assay for detecting P. knowlesi infection. Study participants included 73 uncomplicated malaria patients with PCR species confirmation: 50 P. knowlesi, 20 P. falciparum and 3 P. vivax. Nineteen malaria-negative, non-endemic area controls were also included. The sensitivity of the Eiken Loopamp™ MALARIA Pan Detection kit (Pan LAMP) for detecting each Plasmodium species was evaluated. Sensitivity and specificity of the Eiken Loopamp™ MALARIA Pf Detection kit (Pf LAMP) for P. falciparum were also determined. The limit of detection for each LAMP assay was evaluated, with results compared to PCR. All P. knowlesi patients were also tested by CareStart™ (Pf/VOM) and OptiMAL-IT™ (Pan/Pf) RDTs. The sensitivity of the Pan LAMP assay was 100% for P. knowlesi (95% CI 92.9–100), P. falciparum (95% CI 83.2–100), and P. vivax (95% CI 29.2–100). The Pf LAMP was 100% sensitive and specific for P. falciparum detection, with all P. knowlesi samples having a negative reaction. LAMP sensitivity was superior to both RDTs, with only 10 and 28% of P. knowlesi samples testing positive to CareStart™ and OptiMAL-IT™, respectively. Limit of detection using the Pan LAMP for both P. knowlesi and P. vivax was 2 parasites/μL, comparable to PCR. For P. falciparum both the Pan LAMP and Pf LAMP demonstrated a limit of detection of 20 parasites/μL. The Eiken Loopamp™ MALARIA Pan Detection kit is sensitive for detection of P. knowlesi in low parasitaemia clinical infections, as well as P. falciparum and P. vivax. However, a P. knowlesi-specific field assay in a simpler format would assist correct species identification and initiation of optimal treatment for all malaria patients.

What makes a sustainability tool valuable, practical and useful in real-world healthcare practice? A mixed-methods study on the development of the Long Term Success Tool in Northwest London.

Abstract: Objectives Although improvement initiatives show benefits to patient care, they often fail to sustain. Models and frameworks exist to address this challenge, but issues with design, clarity and usability have been barriers to use in healthcare settings. This work aimed to collaborate with stakeholders to develop a sustainability tool relevant to people in healthcare settings and practical for use in improvement initiatives. Design Tool development was conducted in six stages. A scoping literature review, group discussions and a stakeholder engagement event explored literature findings and their resonance with stakeholders in healthcare settings. Interviews, small-scale trialling and piloting explored the design and tested the practicality of the tool in improvement initiatives. Setting National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care for Northwest London (CLAHRC NWL). Participants CLAHRC NWL improvement initiative teams and staff. Results The iterative design process and engagement of stakeholders informed the articulation of the sustainability factors identified from the literature and guided tool design for practical application. Key iterations of factors and tool design are discussed. From the development process, the Long Term Success Tool (LTST) has been designed. The Tool supports those implementing improvements to reflect on 12 sustainability factors to identify risks to increase chances of achieving sustainability over time. The Tool is designed to provide a platform for improvement teams to share their own views on sustainability as well as learn about the different views held within their team to prompt discussion and actions. Conclusion The development of the LTST has reinforced the importance of working with stakeholders to design strategies which respond to their needs and preferences and can practically be implemented in real-world settings. Further research is required to study the use and effectiveness of the tool in practice and assess engagement with the method over time.

A New Handheld Device for the Detection of Falsified Medicines: Demonstration on Falsified Artemisinin-Based Therapies from the Field.

Abstract: Poor-quality medicines are a major problem for health-care systems in resource-poor settings as identifying falsified medicines requires a complex laboratory infrastructure such as a Medicines Quality Control Laboratory. We report here an evaluation of a low-cost, handheld near-infrared spectrometer (NIRS) device by analyzing a library of artemisinin-based combination therapy (ACT) medicines to determine its usefulness as a drug-screening tool. The "SCiO" research prototype device was used to collect NIR spectra of a library of ACT and artesunate monotherapy medicine samples previously collected in Bioko Island and Equatorial Guinea and Kintampo, Ghana. The quality of these samples had been categorized as falsified, substandard, and quality assured based on the amount of stated active pharmaceutical ingredients detected using high-performance liquid chromatography photodiode array. Numerical analyses were performed on the NIR spectra to assess the usefulness of NIR to identify falsified and substandard medicines. The NIRS device was successful at detecting falsified medicines in all cases where the library contained both quality assured and falsified medicines of the same stated brand of medicines. The NIRS device was successful at identifying substandard amounts of artesunate but not amodiaquine in the ACT samples (N = 15) of artesunate-amodiaquine. This work reveals that this low-cost, portable NIRS device is promising for screening ACTs for falsified samples and could enable widespread drug screening at all points of the health system.

Identification of factors that support successful implementation of care bundles in the acute medical setting: a qualitative study

Abstract: Clinical guidelines offer an accessible synthesis of the best evidence of effectiveness of interventions, providing recommendations and standards for clinical practice. Many guidelines are relevant to the diagnosis and management of the acutely unwell patient during the first 24–48 h of admission. Care bundles are comprised of a small number of evidence-based interventions that when implemented together aim to achieve better outcomes than when implemented individually. Care bundles that are explicitly developed from guidelines to provide a set of related evidence-based actions have been shown to improve the care of many conditions in emergency, acute and critical care settings. This study aimed to review the implementation of two distinct care bundles in the acute medical setting and identify the factors that supported successful implementation. Two initiatives that had used a systematic approach to quality improvement to successfully implement care bundles within the acute medical setting were selected as case studies. Contemporaneous data generated during the initiatives included the review reports, review minutes and audio recordings of the review meetings at different time points. Data were subject to deductive analysis using three domains of the Consolidated Framework for Implementation Research to identify factors that were important in the implementation of the care bundles. Several factors were identified that directly influenced the implementation of the care bundles. Firstly, the availability of resources to support initiatives, which included training to develop quality improvement skills within the team and building capacity within the organisation more generally. Secondly, the perceived sustainability of changes by stakeholders influenced the embedding new care processes into existing clinical systems, maximising their chance of being sustained. Thirdly, senior leadership support was seen as critical not just in supporting implementation but also in sustaining longer-term changes brought about by the initiative. Lastly, practitioner incentives were identified as potential levers to engage junior doctors, a crucial part of the acute medical work force and essential to the initiatives, as there is currently little recognition or reward for involvement The factors identified have been shown to be supportive in the successful implementation of care bundles as a mechanism for implementing clinical guidelines. Addressing these factors at a practitioner and organisational level, alongside the use of a systematic quality improvement approach, should increase the likelihood that care bundles will be implemented successfully to deliver evidence based changes in the acute medical setting.

How to attribute causality in quality improvement: lessons from epidemiology.

Abstract: Quality improvement and implementation (QI&I) initiatives face critical challenges in an era of evidence-based, value-driven patient care. Whether front-line staff, large organisations or government bodies design and run QI&I, there is increasing need to demonstrate impact to justify investment of time and resources in implementing and scaling up an intervention. Decisions about sustaining, scaling up and spreading an initiative can be informed by evidence of causation and the estimated attributable effect of an intervention on observed outcomes. Achieving this in healthcare can be challenging, where interventions often are multimodal and applied in complex systems.1 Where there is weak evidence of causation, credibility in the effectiveness of the intervention is reduced with a resultant reduced desire to replicate. The greater confidence of a causal relationship between QI&I interventions and observed results, the greater our confidence that improvement will result when the intervention occurs in different settings. Guidance exists for design, conduct, evaluation and reporting of QII2–4; the Standards for QUality Improvement Reporting Excellence (SQUIRE) and the Standards for Reporting Implementation Studies (STARI) guidelines were developed specifically for reporting QI&I initiatives.5 6 However, much of this guidance is targeted at larger formal evaluations, and may require levels of resource or expertise not available to all QI&I initiatives. This paper proposes QI&I initiatives, regardless of scope and resources, can be enhanced by applying epidemiological principles, adapted from those promulgated by Austin Bradford Hill.7 Hill proposed nine ‘aspects of association’ that could be considered before ‘… deciding that the most likely interpretation is causation ’.7 His objective was to improve the ability to form scientific judgements about causality. The nine aspects, subsequently referred to as the ‘Bradford Hill Criteria’ (BHC), are considered in the following sections. With roots in causes of disease, the BHC have natural alignment with healthcare. …

Cascading effect in bioprocessing—The impact of mild hypothermia on CHO cell behavior and host cell protein composition

Abstract: A major challenge in downstream purification of monoclonal antibodies (mAb) is the removal of host cell proteins (HCPs). Previous studies have shown that cell culture conditions significantly impact the HCP content at harvest. However, it is currently unclear how process conditions affect physiological changes in the host cell population, and how these changes, in turn, cascade down to change the HCP profile. We examined how temperature downshift (TDS) to mild hypothermia affects key upstream performance indicators, that is antibody titre, HCP concentration and HCP species, across the cell culture decline phase and at harvest through the lens of changes in cellular behavior. Mild hypothermic conditions introduced on day 5 of fed-batch Chinese hamster ovary (CHO) cell bioreactors resulted in a lower cell proliferation rate but larger percentages of healthier cells across the cell culture decline phase compared to bioreactors maintained at standard physiological temperature. Moreover, the onset of apoptosis was less evident in mild hypothermic cultures. Consequently, mild hypothermic cultures took an extra 5 days to reach an integral viable cell concentration (IVCC) and antibody yield similar to that of the control at standard physiological temperature. When cell viability dropped below 80%, mild hypothermic cell cultures had a reduced variety of HCP species by 36%, including approximately 44% and 27% lower proteases and chaperones, respectively, despite having similar HCP concentration. This study suggests that TDS may be a good strategy to provide cleaner downstream feedstocks by reducing the variety of HCPs and to maintain product integrity by reducing the number of proteases and chaperones.

A Randomized, Crossover Trial Evaluating Patient Handling, Preference, and Ease of Use of the Fluticasone Propionate/Formoterol Breath-Triggered Inhaler.

Abstract: Background: Appropriate inhaler selection is of fundamental importance in obstructive lung disease management. Key factors in device selection include a patient's capacity to operate a particular device and their preference for it. Methods: This randomized, open-label, two-period, crossover study (NCT01739387) compared the ability of adolescent and adult patients with obstructive lung disease to correctly handle the fluticasone propionate/formoterol fumarate (FP/FORM; Flutiform®) pressurized metered-dose inhaler (pMDI) and FP/FORM K-haler®, a novel breath-triggered inhaler (BTI), following a simple, standardized training regimen. The primary endpoint was the ability to perform all steps correctly at the first attempt. Secondary endpoints included the ability to perform all critical steps correctly at the first attempt, the requisite number of attempts to successfully use the inhaler, the ability to be trained within 15 minutes, and the ability to trigger the K-haler BTI to actuate at the first at...

Investigation of the degree of organisational influence on patient experience scores in acute medical admission units in all acute hospitals in England using multilevel hierarchical regression modelling.

Abstract: Objectives Previous studies found that hospital and specialty have limited influence on patient experience scores, and patient level factors are more important This could be due to heterogeneity of experience delivery across subunits within organisations We aimed to determine whether organisation level factors have greater impact if scores for the same subspecialty microsystem are analysed in each hospital Setting Acute medical admission units in all NHS Acute Trusts in England Participants We analysed patient experience data from the English Adult Inpatient Survey which is administered to 850 patients annually in each acute NHS Trusts in England We selected all 8753 patients who returned the survey and who were emergency medical admissions and stayed in their admission unit for 1–2 nights, so as to isolate the experience delivered during the acute admission process Primary and secondary outcome measures We used multilevel logistic regression to determine the apportioned influence of host organisation and of organisation level factors (size and teaching status), and patient level factors (demographics, presence of long-term conditions and disabilities) We selected ‘being treated with respect and dignity’ and ‘pain control’ as primary outcome parameters Other Picker Domain question scores were analysed as secondary parameters Results The proportion of overall variance attributable at organisational level was small; 05% (NS) for respect and dignity, 04% (NS) for pain control Long-standing conditions and consequent disabilities were associated with low scores Other item scores also showed that most influence was from patient level factors Conclusions When a single microsystem, the acute medical admission process, is isolated, variance in experience scores is mainly explainable by patient level factors with limited organisational level influence This has implications for the use of generic patient experience surveys for comparison between Trusts and should prompt further research to explore if more discriminant surveys can be developed

Systemic bioavailability (BA) and pharmacodynamics (PD) of fluticasone propionate/formoterol (FP/FORM) via pressurised metered-dose inhaler (pMDI) or a novel breath-triggered inhaler (BTI)

Abstract: Background: Two-stage (S1, S2) adaptive design study to compare systemic BA and PD of FP and FORM via the BTI (Flutiform® K-haler®), and pMDI with (+S) or without a spacer (EudraCT: 2013-000045-39). Methods: Single-dose, randomised, open-label crossover study in healthy adults. S1 assessed pharmacokinetics (PK): 3 period, 3 treatment (FP/FORM 125/5µg 2puffs via BTI, pMDI or pMDI+S). Primary endpoints: AUCt and Cmax. Primary comparisons: for FP, BTI vs pMDI+S; for FORM, BTI vs pMDI (safety confirmed if upper limits of 94.12% confidence intervals [CI] of ratios ≤125%). S2, conducted if S1 primary comparison CI >125%, assessed FORM PD (given S1 outcome): 5 period, 5 treatment (FP/FORM 125/5µg 12puffs via BTI, pMDI or pMDI+S; FP/FORM 125/5µg 4puffs via pMDI; FORM 12µg 5puffs via pMDI). Primary endpoint: maximum reduction in serum potassium within 4h post‑dose. Equivalence: 95% CI for BTI vs pMDI+S and pMDI ratios within 0.5-2. Blood samples were taken up to 36h (S1) or 6h (S2). Results: S1: 43/48 subjects completed. Upper limit (UL) of CIs for FP via BTI vs pMDI+S: AUCt 91.4%, Cmax 87.1%; for FORM via BTI vs pMDI: AUCt 124.1%, Cmax 126.0%. S2: 30/35 subjects completed. The 95% CIs for serum potassium were: BTI vs pMDI+S 0.73-1.28; BTI vs pMDI 0.43-1.46. Conclusion: Systemic exposure of FP via BTI was no greater than pMDI+S; FORM Cmax marginally exceeded 125% UL (S1). S2 confirmed FORM-related systemic safety was similar via BTI, pMDI+S and pMDI. These data support the safety of the Flutiform K-haler product. Sponsor: Mundipharma Research Ltd

Prototype Positive Control Wells for Malaria Rapid Diagnostic Tests: Prospective Evaluation of Implementation Among Health Workers in Lao People's Democratic Republic and Uganda.

Abstract: Rapid diagnostic tests (RDTs) are widely used for malaria diagnosis, but lack of quality control at point of care restricts trust in test results. Prototype positive control wells (PCW) containing recombinant malaria antigens have been developed to identify poor-quality RDT lots. This study assessed community and facility health workers' (HW) ability to use PCWs to detect degraded RDTs, the impact of PCW availability on RDT use and prescribing, and preferred strategies for implementation in Lao People's Democratic Republic (Laos) and Uganda. A total of 557 HWs participated in Laos (267) and Uganda (290). After training, most (88% to ≥ 99%) participants correctly performed the six key individual PCW steps; performance was generally maintained during the 6-month study period. Nearly all (97%) reported a correct action based on PCW use at routine work sites. In Uganda, where data for 127,775 individual patients were available, PCW introduction in health facilities was followed by a decrease in antimalarial prescribing for RDT-negative patients ≥ 5 years of age (4.7-1.9%); among community-based HWs, the decrease was 12.2% (P < 0.05) for all patients. Qualitative data revealed PCWs as a way to confirm RDT quality and restore confidence in RDT results. HWs in malaria-endemic areas are able to use prototype PCWs for quality control of malaria RDTs. PCW availability can improve HWs' confidence in RDT results, and benefit malaria diagnostic programs. Lessons learned from this study may be valuable for introduction of other point-of-care diagnostic and quality-control tools. Future work should evaluate longer term impacts of PCWs on patient management.

Relative pulmonary bioavailability (BA) of fluticasone propionate/formoterol (FP/FORM) via pressurised metered-dose inhaler (pMDI) and a novel breath-triggered inhaler (BTI)

Abstract: Background: This study aimed to show that pulmonary BA of FP and FORM via the BTI (Flutiform® K‑haler®) was no less than via pMDI without a spacer, to establish a bridge from the efficacy of the pMDI to the BTI (EudraCT: 2012-003728-19). Methods: Single-dose, randomised, open-label, 3 treatment, 3 period, cross-over study in healthy adults. Treatments were FP/FORM 125/5µg, 2 puffs via BTI, pMDI or pMDI with spacer (pMDI+S). To measure pulmonary BA only, all subjects ingested charcoal suspension pre-dose and at 15min and 1h post-dose. The primary objective was to compare relative BA of FP and FORM via BTI and pMDI (secondary objectives included BTI vs pMDI+S). Acceptable pulmonary BA was concluded if the lower limit of the confidence intervals (CIs) of the BAI/pMDI ratios were ≥80%. Blood samples were taken pre-dose and for 36h post-dose. Results: Of 47 subjects randomised, 45 (95.7%) completed. The AUCt and Cmax ratios of BTI:pMDI for both FP and FORM analytes were between 107-131%. As the lower limits of the CIs of all BTI:pMDI ratios were >80% the study goal was met (Table). Conclusion: Pulmonary BA of FP/FORM via BTI was no less than via pMDI without spacer, indicating that the Flutiform K-haler product would be at least as effective as pMDI. Sponsor: Mundipharma Research Ltd Table. BA comparisons

Isqua17-3279how many hospital websites provide information to attract patients to attend cardiac/pulmonary rehabilitation across england?

Abstract: were carried out to understand what is important to patients, how the concept of Always Events might apply to the New Zealand context, and how the lower scoring areas of the national adult inpatient experience survey could be improved over time. The piloting phase closely followed the protocol for rolling out an Always Event. Always Events are defined as aspects of the patient experience that are so important to patients, care partners, and service users that health care providers must aim to perform them consistently for every individual, every time. The ‘event’ is best measured by linking it to the patient experience surveys, particularly the lower scoring areas. This offers an opportunity to improve as well as measure impact. Results: Early findings suggest that a systematic approach to capturing the patient experience as it related to the lower scoring areas of the national adult inpatient experience survey provide an opportunity to improve patient care that the national aggregated results do not offer. Specific patient experience data can be used to tailor specific interventions and provide parameters for a national quality and safety marker. Conclusion: As a result of the pilot, a quality and safety marker for patient experience can now be investigated. The challenge remains to adapt interventions across all services that are truly patientcentered and informed by the patient voice. Reference http://www.ihi.org/resources/Pages/Tools/Always-Events-Toolkit.aspx (accessed 9 February 2017).

Prevention through Learning: working together to drive high-quality care.

Abstract: in may 2015 the academy of medical royal Colleges and faculties in scotland published a report on Learning from Serious Failings in Care.1 this reviewed the reports on failings experienced in mid staffordshire, lanarkshire, vale of leven and aberdeen and made a series of recommendations regarding how the nhs in scotland could learn from these failings and prevent recurrence. it became clear that poor team working and engagement between managers and clinicians had been recurring factors and contributed to these failings. recognising the need to consider how we could best collectively address these defi ciencies, the scottish academy convened a meeting in october 2015. invited stakeholders included nhs board Chairs, Chief executives, medical directors, representatives of the royal Colleges and other related medical and nursing organisations including nhs healthcare improvement scotland and rCn scotland. the report and evening meeting provided a catalyst for change and for progressing our aim of delivering the highest possible quality of care. reflecting the need for managers and clinicians to work more effectively together, and our desire to lead by example, the scottish academy partnered with the scottish institute of health management and agreed to convene two events with the support of the good governance institute – the fi rst in a series of workshops for nhs board non-executive directors (31 october 2016) and a related national stakeholder meeting (11 november 2016). this paper summarises the national stakeholder meeting.