Screening mammography aims to identify breast cancer at earlier stages of the disease, when treatment can be more successful1. Despite the existence of screening programmes worldwide, the interpretation of mammograms is affected by high rates of false positives and false negatives2. Here we present an artificial intelligence (AI) system that is capable of surpassing human experts in breast cancer prediction. To assess its performance in the clinical setting, we curated a large representative dataset from the UK and a large enriched dataset from the USA. We show an absolute reduction of 5.7% and 1.2% (USA and UK) in false positives and 9.4% and 2.7% in false negatives. We provide evidence of the ability of the system to generalize from the UK to the USA. In an independent study of six radiologists, the AI system outperformed all of the human readers: the area under the receiver operating characteristic curve (AUC-ROC) for the AI system was greater than the AUC-ROC for the average radiologist by an absolute margin of 11.5%. We ran a simulation in which the AI system participated in the double-reading process that is used in the UK, and found that the AI system maintained non-inferior performance and reduced the workload of the second reader by 88%. This robust assessment of the AI system paves the way for clinical trials to improve the accuracy and efficiency of breast cancer screening. An artificial intelligence (AI) system performs as well as or better than radiologists at detecting breast cancer from mammograms, and using a combination of AI and human inputs could help to improve screening efficiency.

/pdf/international-evaluation-of-an-ai-system-for-breast-cancer-4c1yg7x284.pdf

International evaluation of an AI system for breast cancer screening.

Markov models are often employed to represent stochastic processes, that is, random processes that evolve over time. In a healthcare context, Markov models are particularly suited to modelling chronic disease. In this article, we describe the use of Markov models for economic evaluation of healthcare interventions. The intuitive way in which Markov models can handle both costs and outcomes make them a powerful tool for economic evaluation modelling. The time component of Markov models can offer advantages of standard decision tree models, particularly with respect to discounting. This paper gives a comprehensive description of Markov modelling for economic evaluation, including a discussion of the assumptions on which the type of model is based, most notably the memoryless quality of Markov models often termed the ‘Markovian assumption’. A hypothetical example of a drug intervention to slow the progression of a chronic disease is employed to demonstrate the modelling technique and the possible methods of analysing Markov models are explored. Analysts should be aware of the limitations of Markov models, particularly the Markovian assumption, although the adept modeller will often find ways around this problem.

An introduction to markov modelling for economic evaluation

A version of this guideline for pregnant women, their partners and the public, called Routine antenatal care for healthy pregnant women. Understanding NICE guidance: information for pregnant women, their families and the public, is also available (reproduced as Appendix 1 in this version). It can be downloaded from the NICE website (www.nice.org.uk) or ordered from the NHS Response Line (087

Antenatal care : routine care for the healthy pregnant woman

PURPOSE: To determine the false-negative rate in screening mammography, the capability of computer-aided detection (CAD) to identify these missed lesions, and whether or not CAD increases the radiologists' recall rate. MATERIALS AND METHODS: All available screening mammograms that led to the detection of biopsy-proved cancer (n = 1,083) and the most recent corresponding prior mammograms (n = 427) were collected from 13 facilities. Panels of radiologists evaluated the retrospectively visible prior mammograms by means of blinded review. All mammograms were analyzed by a CAD system that marks features associated with cancer. The recall rates of 14 radiologists were prospectively measured before and after installation of the CAD system. RESULTS: At retrospective review, 67% (286 of 427) of screening mammography–detected breast cancers were visible on the prior mammograms. At independent, blinded review by panels of radiologists, 27% (115 of 427) were interpreted as warranting recall on the basis of a statisti...

Potential contribution of computer-aided detection to the sensitivity of screening mammography.

Breast cancer is the second-most common and leading cause of cancer death among women. It has become a major health issue in the world over the past 50 years, and its incidence has increased in recent years. Early detection is an effective way to diagnose and manage breast cancer. Computer-aided detection or diagnosis (CAD) systems can play a key role in the early detection of breast cancer and can reduce the death rate among women with breast cancer. The purpose of this paper is to provide an overview of recent advances in the development of CAD systems and related techniques. We begin with a brief introduction to some basic concepts related to breast cancer detection and diagnosis. We then focus on key CAD techniques developed recently for breast cancer, including detection of calcifications, detection of masses, detection of architectural distortion, detection of bilateral asymmetry, image enhancement, and image retrieval.

Computer-Aided Detection and Diagnosis of Breast Cancer With Mammography: Recent Advances

OBJECTIVES: To provide evidence to inform policy decisions about the most appropriate newborn screening strategy for congenital heart defects, identifying priorities for future research that might reduce important uncertainties in the evidence base for such decisions DATA SOURCES: Electronic databases Groups of parents and health professionals REVIEW METHODS: A systematic review of the published medical literature concerning outcomes for children with congenital heart defects was carried out A decision analytic model was developed to assess the cost-effectiveness of alternative screening strategies for congenital heart defects relevant to the UK A further study was then carried out using a self-administered anonymous questionnaire to explore the perspectives of parents and health professionals towards the quality of life of children with congenital heart defects The findings from a structured review of the medical literature regarding parental experiences were linked with those from a focus group of parents of children with congenital heart defects RESULTS: Current newborn screening policy comprises a clinical examination at birth and 6 weeks, with specific cardiac investigations for specified high-risk children Routine data are lacking, but under half of affected babies, not previously identified antenatally or because of symptoms, are identified by current newborn screening There is evidence that screen-positive infants do not receive timely management Pulse oximetry and echocardiography, in addition to clinical examination, are alternative newborn screening strategies but their cost-effectiveness has not been adequately evaluated in a UK setting In a population of 100,000 live-born infants, the model predicts 121 infants with life-threatening congenital heart defects undiagnosed at screening, of whom 82 (68%) and 83 (69%) are detected by pulse oximetry and screening echocardiography, respectively, but only 39 (32%) by clinical examination alone Of these, 71, 71 and 34, respectively, receive a timely diagnosis The model predicts 46 (05%) false-positive screening diagnoses per 100,000 infants with clinical examination, 1168 (13%) with pulse oximetry and 4857 (54%) with screening echocardiography The latter includes infants with clinically non-significant defects Total programme costs are predicted of pound 300,000 for clinical examination, pound 480,000 for pulse oximetry and pound 354 million for screening echocardiography The additional cost per additional timely diagnosis of life-threatening congenital heart defects ranges from pound 4900 for pulse oximetry to pound 45 million for screening echocardiography Including clinically significant congenital heart defects gives an additional cost per additional diagnosis of pound 1500 for pulse oximetry and pound 36,000 for screening echocardiography Key determinants for cost-effectiveness are detection rates for pulse oximetry and screening echocardiography Parents and health professionals place similar values on the quality of life outcomes of children with congenital heart defects and both are more averse to neurological than to cardiac disability Adverse psychosocial effects for parents are focused around poor management and/or false test results CONCLUSIONS: Early detection through newborn screening potentially can improve the outcome of congenital heart defects; however the current programme performs poorly, and lacks monitoring of quality assurance, performance management and longer term outcomes Pulse oximetry is a promising alternative newborn screening strategy but further evaluation is needed to obtain more precise estimates of test performance and to inform optimal timing, diagnostic and management strategies Although screening echocardiography is associated with the highest detection rate, it is the most costly strategy and has a 5% false-positive rate Improving antenatal detection of congenital heart defects increases the cost per timely postnatal diagnosis afforded by any newborn screening strategy but does not alter the relative effects of the strategies An improvement of timely management of screen positive infants is essential Further research is required to refine the detection rate and other aspects of pulse oximetry, to evaluate antenatal screening strategies more directly, and to investigate the psychosocial effects of newborn screening for congenital heart defects

Newborn screening for congenital heart defects: a systematic review and cost-effectiveness analysis.

/pdf/antenatal-and-neonatal-haemoglobinopathy-screening-in-the-uk-gjqz37uhzd.pdf

Antenatal and neonatal haemoglobinopathy screening in the UK: review and economic analysis.

Objective: To compare mammography reading by one radiologist with independent reading by two radiologists. Design: An observational non-randomised trial at St Margaret9s Hospital, Epping. Subjects: 33734 consecutive attenders for breast screening in the main trial and a sample of 132 attenders for assessment who provided data on private costs. Interventions: Three reporting policies were compared: single reading, consensus double reading, and non-consensus double reading. Main outcome measures: Numbers of cancers detected, recall rates, screening and assessment costs, and cost effectiveness ratios. Results: A policy of double reading followed by consensus detected an additional nine cancers per 10000 women screened (95% confidence interval 5 to 13) compared with single reading. A non-consensus double reading policy detected an additional 10 cancers per 10000 women screened (95% confidence interval 6 to 14). The difference in numbers of cancers detected between the consensus and non-consensus double reading policies was not significant (95% confidence interval -0.2 to 2.2). The proportion of women recalled for assessment after consensus double reading was significantly lower than after single reading (difference 2.7%; 95% confidence interval 2.4% to 3.0%). The recall rate with the non-consensus policy was significantly higher than with single reading (difference 3.0%; 2.5% to 3.5%). Consensus double reading cost less than single reading (saving pounds sterling4853 per 10000 women screened). Non-consensus double reading cost more than single reading (difference pounds sterling19259 per 10000 women screened). Conclusions: In the screening unit studied a consensus double reading policy was more effective and less costly than a single reading policy. Key messages Key messages Double reading of screening mammograms detects more cancers than does single reading Double reading with consensus reduces recall rates and has a lower total cost than single reading Breast screening units should consider adopting consensus double reporting for the first screening examination in order to improve efficiency Double reading with consensus is also likely to confer benefits at subsequent screening examinations, though the magnitude and cost effectiveness of these benefits are not known

/pdf/mammography-screening-an-incremental-cost-effectiveness-3f59jmsfit.pdf

Mammography screening : an incremental cost effectiveness analysis of double versus single reading of mammograms

The increased use of modelling techniques as a methodological tool in the economic evaluation of health care technologies has, in the main, been limited to two approaches – decision trees and Markov chain models. The former are suited to modelling simple scenarios that occur over a short time period, whilst Markov chain models allow longer time periods to be modelled, in continuous time, where the timing of an event is uncertain. In the context of economic evaluation, a less well developed technique is discrete event simulation, which may allow even greater flexibility.

Selecting a decision model for economic evaluation: a case study and review.

Providers and purchasers of health care are increasingly looking to the results of economic evaluations for guidance when making their decisions. In this paper the authors argue that there are dangers involved in the naive and unthinking use of published cost-effectiveness information outside the setting in which the information was generated. In considering whether the results of a published study are likely to be relevant locally, decision-makers are encouraged to assess whether the values of the key parameters reported in the published study apply locally. The possible sources of variation are described: unit cost differences; differences in the prevalence, incidence or natural history of disease; and differences in the comparators. In situations where the only source of variation is that local unit costs are different, local values can be substituted in the published analysis and local cost-effectiveness results estimated. Where the other sources of variation exist, the decision-maker is required to m...

Jackie Brown

Papers

Newborn screening for congenital heart defects: a systematic review and cost-effectiveness analysis.

Antenatal and neonatal haemoglobinopathy screening in the UK: review and economic analysis.

Mammography screening : an incremental cost effectiveness analysis of double versus single reading of mammograms

Selecting a decision model for economic evaluation: a case study and review.

Extrapolation of cost-effectiveness information to local settings.