An Interpretable and Accurate Deep-Learning Diagnosis Framework Modeled With Fully and Semi-Supervised Reciprocal Learning

The deployment of automated deep-learning classifiers in clinical practice has the potential to streamline the diagnosis process and improve the diagnosis accuracy, but the acceptance of those classifiers relies on both their accuracy and interpretability. In general, accurate deep-learning classifiers provide little model interpretability, while interpretable models do not have competitive classification accuracy. In this paper, we introduce a new deep-learning diagnosis framework, called InterNRL, that is designed to be highly accurate and interpretable. InterNRL consists of a student-teacher framework, where the student model is an interpretable prototype-based classifier (ProtoPNet) and the teacher is an accurate global image classifier (GlobalNet). The two classifiers are mutually optimised with a novel reciprocal learning paradigm in which the student ProtoPNet learns from optimal pseudo labels produced by the teacher GlobalNet, while GlobalNet learns from ProtoPNet's classification performance and pseudo labels. This reciprocal learning paradigm enables InterNRL to be flexibly optimised under both fully- and semi-supervised learning scenarios, reaching state-of-the-art classification performance in both scenarios for the tasks of breast cancer and retinal disease diagnosis. Moreover, relying on weakly-labelled training images, InterNRL also achieves superior breast cancer localisation and brain tumour segmentation results than other competing methods.

Test-set training improves the detection rates of invasive cancer in screening mammography

AIM

To investigate if mammographic test-set participation affects routine breast cancer screening performance.

MATERIALS AND METHODS

Clinical audit data between 2008 and 2018 were collected for 35 breast screen readers who participated in the BreastScreen Reader Assessment Strategy (BREAST) and 22 readers with no history of test-set participation. For BREAST readers, the annual audit data were divided according to the year they completed their first test set, and the same years were used randomly to align and divide the data of non-BREAST readers into pre- and post-test set periods. Multiple audit parameters were inspected retrospectively for the two cohorts to identify how their reading performance has evolved in screening mammography.

RESULTS

Investigating 2 calendar years before and after test-set participation, BREAST and non-BREAST readers recalled lower rates of women in the latter period (p=0.03 and p=0.02, respectively). They also improved their positive predictive value (PPV; p=0.01 and p=0.02, respectively). BREAST readers additionally improved their detection rates of invasive cancer (p=0.02) and all cancers (p=0.01). In an extended 3-year comparison, similar improvements occurred in the recall rate for BREAST (p=0.02) and non-BREAST readers (p=0.02) and in PPV (p=0.001, 0.01, respectively); however, improvements in detection rates also occurred exclusively in BREAST readers' performance for invasive cancer (p=0.04), DCIS (p=0.05), and all cancers (p=0.02); however, significant improvements in detection did not involve <15 mm invasive cancers in both periods. Meanwhile, non-BREAST readers demonstrated a decrease in sensitivity (p=0.02).

CONCLUSION

Participation in test sets is linked to over-time improvements in most audit-measured cancer detection rates.

Utopia versus dystopia: Professional perspectives on the impact of healthcare artificial intelligence on clinical roles and skills

BACKGROUND

Alongside the promise of improving clinical work, advances in healthcare artificial intelligence (AI) raise concerns about the risk of deskilling clinicians. This purpose of this study is to examine the issue of deskilling from the perspective of diverse group of professional stakeholders with knowledge and/or experiences in the development, deployment and regulation of healthcare AI.

METHODS

We conducted qualitative, semi-structured interviews with 72 professionals with AI expertise and/or professional or clinical expertise who were involved in development, deployment and/or regulation of healthcare AI. Data analysis using combined constructivist grounded theory and framework approach was performed concurrently with data collection.

FINDINGS

Our analysis showed participants had diverse views on three contentious issues regarding AI and deskilling. The first involved competing views about the proper extent of AI-enabled automation in healthcare work, and which clinical tasks should or should not be automated. We identified a cluster of characteristics of tasks that were considered more suitable for automation. The second involved expectations about the impact of AI on clinical skills, and whether AI-enabled automation would lead to worse or better quality of healthcare. The third tension implicitly contrasted two models of healthcare work: a human-centric model and a technology-centric model. These models assumed different values and priorities for healthcare work and its relationship to AI-enabled automation.

CONCLUSION

Our study shows that a diverse group of professional stakeholders involved in healthcare AI development, acquisition, deployment and regulation are attentive to the potential impact of healthcare AI on clinical skills, but have different views about the nature and valence (positive or negative) of this impact. Detailed engagement with different types of professional stakeholders allowed us to identify relevant concepts and values that could guide decisions about AI algorithm development and deployment.

Australian women's judgements about using artificial intelligence to read mammograms in breast cancer screening

Objective

Mammographic screening for breast cancer is an early use case for artificial intelligence (AI) in healthcare. This is an active area of research, mostly focused on the development and evaluation of individual algorithms. A growing normative literature argues that AI systems should reflect human values, but it is unclear what this requires in specific AI implementation scenarios. Our objective was to understand women's values regarding the use of AI to read mammograms in breast cancer screening.

Methods

We ran eight online discussion groups with a total of 50 women, focused on their expectations and normative judgements regarding the use of AI in breast screening.

Results

Although women were positive about the potential of breast screening AI, they argued strongly that humans must remain as central actors in breast screening systems and consistently expressed high expectations of the performance of breast screening AI. Women expected clear lines of responsibility for decision-making, to be able to contest decisions, and for AI to perform equally well for all programme participants. Women often imagined both that AI might replace radiographers and that AI implementation might allow more women to be screened: screening programmes will need to communicate carefully about these issues.

Conclusions

To meet women's expectations, screening programmes should delay implementation until there is strong evidence that the use of AI systems improves screening performance, should ensure that human expertise and responsibility remain central in screening programmes, and should avoid using AI in ways that exacerbate inequities.

Exploration of potential indicators of burnout, psychological distress and post-traumatic stress disorder, among Australian female first responders

There is limited research on the psychological wellbeing of female first responders (FRs) and therefore we explore potential indicators of burnout, psychological distress and post-traumatic stress disorder among Australian female FRs. We conducted an online health survey among Australian female FRs (fire, police, paramedical, aeromedical, remote area and other e.g., State Emergency Service). Of the 422 eligible participants who submitted the online survey, 286 completed at least 80% of all survey questions and were used in the final analyses. The main outcomes of interest were moderate burnout (compared to low burnout) and high scores for combined PCL-5/K10 (compared to low scores). Using logistical regression stepwise regression models, we analysed associations between the outcomes of interest and various work-psychosocial factors. Results showed the strongest indicators of moderate burnout to be, 1) returning to work with <12-hour break, 2) exposure to gossip and slander, 3) not enough time to do things, 4) and having experienced rape/sexual assault. The strongest indicators of higher PCL-5/K10 scores were, 1) exposure to unpleasant teasing, 2) pressure at work and home, 3) having experienced physical violence (e.g., beaten/mugged), and 4) someone close to them died unexpectedly. These findings show workforce stressors have more impact on female FRs psychological wellbeing, compared to lifetime traumatic exposures.

Clinical performance progress of BREAST participants: the impact of test-set participation

AIM

To investigate if positive changes in the clinical performance of radiologists are associated with reading mammographic test sets.

MATERIALS AND METHODS

This study investigated the clinical audit history for a cohort of 39 participants in the BreastScreen Reader Assessment Strategy who have read for BreastScreen New South Wales in the period between 2010 and 2018, inclusively. Based on the year in which each radiologist completed his or her first test set, data of multiple clinical audit metrics from two calendar years before test-set reading were compared against similar data from three different periods after test-set completion. The same process was repeated after dividing radiologists into two subgroups based on their median screen-reading volume (3,688), to test if experience is a determinant of post-test set performance.

RESULTS

On average, radiologists showed significant improvements (p<0.05) in the recall rate for subsequent screening rounds, in positive predictive value 1 (PPV1), and in specificity. When dividing radiologists based on their average annual reading volume, radiologists with higher reading numbers demonstrated similar significant improvements in the recall rate and in PPV1. In addition, they showed significant improvements in the detection rates of invasive breast cancer and ductal carcinoma in situ (DCIS). In contrast, the radiologists with lower reading volume indicated significant changes in the recall rate and in PPV1, both accruing in one of the three compared periods.

CONCLUSION

Mammographic test-set participants improve over time in identifying normal breast screens and detecting breast cancer in association with reading higher volumes of breast screening cases.

Factors influencing the recruitment and retention of female first responders: a mixed methods systematic review protocol

OBJECTIVE

The objective of this review is to produce a set of integrated findings of quantitative and qualitative evidence regarding workplace recruitment and retention factors (including departure) of female first responders to inform recommendations for policy and practice.

INTRODUCTION

Historically, first responder workforces such as police officers, firefighters, search and rescue personnel, medical technicians, and paramedics have been largely male dominated. Over the past few decades, however, there has been a steady increase in the number of women entering this field. However, there is minimal research examining factors that influence the recruitment/retention of female first responders. The intention of this review is to identify barriers and facilitators to recruitment and retention of female first responders and to identify any differences between sexes/genders.

INCLUSION CRITERIA

This review will summarize qualitative and quantitative research examining factors influencing the recruitment/retention of female first responders. Such factors may include job satisfaction, quality of work life, burnout, compassion fatigue, and intent to remain in the workforce.

METHODS

MEDLINE (PubMed), CINAHL (EBSCO), PsycINFO (APA), PTSDpubs (formerly PILOTS; ProQuest), Embase (Elsevier), and Scopus (Elsevier) will be searched for studies published in English from 2009 to the present. Unpublished studies will be searched in Google Scholar, and ProQuest Dissertations and Theses Global. Both quantitative and qualitative studies will be screened for inclusion and critically appraised for methodological quality by two independent reviewers. Both types of data will be extracted using JBI tools for mixed methods systematic reviews. A convergent integrated approach to synthesis and integration will be used.

SYSTEMATIC REVIEW REGISTRATION NUMBER

PROSPERO CRD42020156524.

Improvement of Cancer Detection on Mammograms via BREAST Test Sets

BACKGROUND

Breast Screen Reader Assessment Strategy (BREAST) is an innovative training and research program for radiologists in Australia and New Zealand. The aim of this study is to evaluate the efficacy of BREAST test sets in improving readers' performance in detecting cancers on mammograms.

MATERIALS AND METHODS

Between 2011 and 2018, 50 radiologists (40 fellows, 10 registrars) completed three BREAST test sets and 17 radiologists completed four test sets. Each test set contained 20 biopsy-proven cancer and 40 normal cases. Immediate image-based feedback was available to readers after they completed each test set which allowed the comparison of their selections with the truth. Case specificity, case sensitivity, lesion sensitivity, the Receiver Operating Characteristic (ROC) Area Under the Curve (AUC) and Jackknife Free-Response Receiver Operating Characteristic (JAFROC) Figure of Merit (FOM) were calculated for each reader. Kruskal-Wallis test was utilized to compare scores of the radiologist and registrars across all test-sets whilst Wilcoxon signed rank test was to compare the scores between pairs of test sets.

RESULTS

Significant improvements in lesion sensitivity ranging from 21% to 31% were found in radiologists completing later test sets compared to first test set (p ≤ 0.01). Eighty three percent of radiologists achieved higher performance in lesion sensitivity after they completed the first read. Registrars had significantly better scores in the third test set compared to the first set with mean increases of 79% in lesion sensitivity (p = 0.005) and 37% in JAFROC (p = 0.02). Sixty percent and 100% of registrars increased their scores in lesion sensitivity in the second and third reads compared to the first read while the percentage of registrars with higher scores in JAFROC was 80%.

CONCLUSION

Introduction of BREAST into national training programs appears to have an important impact in promoting diagnostic efficacy amongst radiologists and radiology registrars undergoing mammographic readings.

The ethical, legal and social implications of using artificial intelligence systems in breast cancer care

Breast cancer care is a leading area for development of artificial intelligence (AI), with applications including screening and diagnosis, risk calculation, prognostication and clinical decision-support, management planning, and precision medicine. We review the ethical, legal and social implications of these developments. We consider the values encoded in algorithms, the need to evaluate outcomes, and issues of bias and transferability, data ownership, confidentiality and consent, and legal, moral and professional responsibility. We consider potential effects for patients, including on trust in healthcare, and provide some social science explanations for the apparent rush to implement AI solutions. We conclude by anticipating future directions for AI in breast cancer care. Stakeholders in healthcare AI should acknowledge that their enterprise is an ethical, legal and social challenge, not just a technical challenge. Taking these challenges seriously will require broad engagement, imposition of conditions on implementation, and pre-emptive systems of oversight to ensure that development does not run ahead of evaluation and deliberation. Once artificial intelligence becomes institutionalised, it may be difficult to reverse: a proactive role for government, regulators and professional groups will help ensure introduction in robust research contexts, and the development of a sound evidence base regarding real-world effectiveness. Detailed public discussion is required to consider what kind of AI is acceptable rather than simply accepting what is offered, thus optimising outcomes for health systems, professionals, society and those receiving care.

Increased COX-2 expression in epithelial and stromal cells of high mammographic density tissues and in a xenograft model of mammographic density

Mammographic density (MD) adjusted for age and body mass index is one of the strongest known risk factors for breast cancer. Given the high attributable risk of MD for breast cancer, chemoprevention with a safe and available agent that reduces MD and breast cancer risk would be beneficial. Cox-2 has been implicated in MD-related breast cancer risk, and was increased in stromal cells in high MD tissues in one study. Our study assessed differential Cox-2 expression in epithelial and stromal cells in paired samples of high and low MD human breast tissue, and in a validated xenograft biochamber model of MD. We also examined the effects of endocrine treatment upon Cox-2 expression in high and low MD tissues in the MD xenograft model. Paired high and low MD human breast tissue samples were immunostained for Cox-2, then assessed for differential expression and staining intensity in epithelial and stromal cells. High and low MD human breast tissues were separately maintained in biochambers in mice treated with Tamoxifen, oestrogen or placebo implants, then assessed for percentage Cox-2 staining in epithelial and stromal cells. Percentage Cox-2 staining was greater for both epithelial (p = 0.01) and stromal cells (p < 0.0001) of high compared with low MD breast tissues. In high MD biochamber tissues, percentage Cox-2 staining was greater in stromal cells of oestrogen-treated versus placebo-treated tissues (p = 0.05).

Topographic position of forelimb motoneuron pools is conserved in vertebrate evolution

The neuromotor conservatism hypothesis predicts that neuromotor patterns in homologous tetrapod muscles are conserved evolutionarily despite the musculoskeletal modifications of vertebrate limbs. A complete description of the anatomical organization of the neurons innervating homologous limb muscles is a prerequisite to any test of the neuromotor conservatism hypothesis. This study uses the retrograde neuronal tracer WGA-HRP to selectively label the motor neuron pools of seven homologous forelimb muscles in mice (Mus musculus) and iguanas (Iguana iguana): Mm. pectoralis, spinodeltoideus, biceps brachii, lateral and long heads of triceps brachii, and the supraspinatus and infraspinatus (in mice) or their reptilian homolog, the supracoracoideus (in iguanas). In vertebrates, motoneurons are arranged in longitudinal columns of cells in the ventral horn of the spinal cord. Mouse motor pools average 1,952 microns in length, except the pectoralis pool which averaged 2,949 microns in length. Iguana pools average 3,196 microns in length. The number of neurons per pool ranged from 70-199 in mice and from 58-114 neurons in iguanas. In both iguanas and mice the motor pools for the spinodeltoids, biceps, and the supracoracoideus (or its mammalian homologs) lie anterior to the pectoralis and triceps motor pools. In the transverse plane, the pectoralis pool lies medial to those of the triceps. The pools of the biceps and spinodeltoids are located dorsal and lateral to those of the pectoralis and supracoracoideus (or its homologs in mammals). The resulting motor pool maps support the hypothesis that the anatomical organization of motoneurons in ancestral reptiles has been retained in these two tetrapod descendents.

Once daily intranasal fluticasone propionate is effective for perennial allergic rhinitis

The efficacy of intranasal fluticasone propionate 200 micrograms once daily or 100 micrograms twice daily in treating perennial allergic rhinitis was evaluated in a randomized, double-blind, placebo-controlled study of 24 weeks' duration in 365 patients. Clinician-rated and patient-rated total nasal symptom severity scores were improved within 1 week of treatment with either regimen of fluticasone propionate and improvement was maintained over the 24-week treatment period. Clinician-rated overall evaluation indicated a significantly better response in the two fluticasone propionate groups compared with the placebo group. All efficacy evaluations indicated no difference in response between the fluticasone propionate 200 micrograms once-daily and 100 micrograms twice-daily groups. Patients in both fluticasone propionate groups had significantly less nasal obstruction upon awakening than the placebo group at all assessment periods. Fewer patients in either fluticasone propionate group used antihistamine rescue medication compared with the placebo group. The percentage of patients with nasal eosinophils and basophils at the end of the 24-week treatment period was significantly lower in both fluticasone propionate groups compared with the placebo group. Safety evaluations indicated that intranasal fluticasone propionate was as safe as placebo when given as 200 micrograms once daily or 100 micrograms twice daily. The incidence of drug-related adverse events was similar among the fluticasone propionate and placebo groups except for the incidence of epistaxis and blood in nasal mucus which was somewhat higher in the fluticasone propionate twice-daily group. There was no changes in the opthalmic examinations to suggest corticosteriod-induced posterior subcapsular cataract formation.(ABSTRACT TRUNCATED AT 250 WORDS)

Current status of imaging indeterminate renal masses

In 1986 Bosniak proposed a classification of cysts and cystic renal masses in an attempt to define precisely the radiologic findings for lesions not requiring further evaluation and for those that should be explored surgically in an attempt to salvage normal renal parenchyma. This classification has been extremely useful in expanding the number of patients treated conservatively. The following entities can be considered benign and require no further evaluation: the classic simple cyst (Bosniak category 1), a cyst with benign calcification (see Fig. 1A and B), a hyperdense cyst (see Fig. 2), and a cyst with thin (less than or equal to 1mm) septations (see Fig. 5A; Bosniak category 2). When any of the following features are present, further evaluation is warranted: aggressive calcifications (see Fig. 1C-E), abnormal density not fulfilling the criteria of a hyperdense cyst (see Figs. 3 and 4), or nodular septations (see Fig. 5B and C), or wall thickening (see Fig. 7; Bosniak) category 3). Small masses that cannot be characterized accurately also require further evaluation. Multiloculated masses(see Fig. 9) require surgery (Bosniak category 3). Most cases of multiple localized renal cysts can be distinguished from multiloculated masses and do not require surgery (see Fig. 10). Evaluation of cyst fluid by MR imaging has not been proved reliable in differentiating benign and malignant masses. MR imaging will become more useful when its spatial resolution equals that of CT. Masses that remain indeterminate after careful evaluation by ultrasonography and CT scanning can be evaluated additionally by angiography or cyst puncture. Management options for persistent enigmatic masses include follow-up radiologic studies, exploration, local excision, and nephrectomy.

Plasma level studies of penbutolol after oral dose in man

Plasma levels of penbutolol (HOE 893d) were determined in eight healthy adult male subjects after oral administration of 50-mg capsules. Fast absorpiton of the drug from the gastrointestinal tract was indicated by the rapid increase in plasma levels during the absorption phase, with a peak time at about 1 hour after dosing in all subjects. After the peak level, plasma concentrations declined biexponentially, with an average half-life of 2.5 and 27 hours for the fast and slow disposition phases, respectively. These values were in good agreement with data previously found for this drug. Cumulative excretion of intact drug in the urine of the eight subjects during 72 hours after dosing was less than 4 per cent, except for one subject who excreted 9.82 per cent of the dose. Large individual variations were found for area under the plasma level curves, disposition rates, and amounts of intact drug excreted in the urine. Significant pharmacologic effects were noted in all eight subjects at the 50-mg dose level, and mild side effects were evident in one half of these subjects. The average drop in blood pressure and pulse rate for all subjects was 26/18 mm Hg and 19 beats per minute, respectively.