Prospective implementation of AI-assisted screen reading to improve early detection of breast cancer.

Artificial intelligence (AI) has the potential to improve breast cancer screening; however, prospective evidence of the safe implementation of AI into real clinical practice is limited. A commercially available AI system was implemented as an additional reader to standard double reading to flag cases for further arbitration review among screened women. Performance was assessed prospectively in three phases: a single-center pilot rollout, a wider multicenter pilot rollout and a full live rollout. The results showed that, compared to double reading, implementing the AI-assisted additional-reader process could achieve 0.7-1.6 additional cancer detection per 1,000 cases, with 0.16-0.30% additional recalls, 0-0.23% unnecessary recalls and a 0.1-1.9% increase in positive predictive value (PPV) after 7-11% additional human reads of AI-flagged cases (equating to 4-6% additional overall reading workload). The majority of cancerous cases detected by the AI-assisted additional-reader process were invasive (83.3%) and small-sized (≤10 mm, 47.0%). This evaluation suggests that using AI as an additional reader can improve the early detection of breast cancer with relevant prognostic features, with minimal to no unnecessary recalls. Although the AI-assisted additional-reader workflow requires additional reads, the higher PPV suggests that it can increase screening effectiveness.

Comparing Prognostic Factors of Cancers Identified by Artificial Intelligence (AI) and Human Readers in Breast Cancer Screening.

Invasiveness status, histological grade, lymph node stage, and tumour size are important prognostic factors for breast cancer survival. This evaluation aims to compare these features for cancers detected by AI and human readers using digital mammography. Women diagnosed with breast cancer between 2009 and 2019 from three UK double-reading sites were included in this retrospective cohort evaluation. Differences in prognostic features of cancers detected by AI and the first human reader (R1) were assessed using chi-square tests, with significance at p < 0.05. From 1718 screen-detected cancers (SDCs) and 293 interval cancers (ICs), AI flagged 85.9% and 31.7%, respectively. R1 detected 90.8% of SDCs and 7.2% of ICs. Of the screen-detected cancers detected by the AI, 82.5% had an invasive component, compared to 81.1% for R1 (p-0.374). For the ICs, this was 91.5% and 93.8% for AI and R1, respectively (p = 0.829). For the invasive tumours, no differences were found for histological grade, tumour size, or lymph node stage. The AI detected more ICs. In summary, no differences in prognostic factors were found comparing SDC and ICs identified by AI or human readers. These findings support a potential role for AI in the double-reading workflow.

Multi-vendor evaluation of artificial intelligence as an independent reader for double reading in breast cancer screening on 275,900 mammograms.

BACKGROUND: Double reading (DR) in screening mammography increases cancer detection and lowers recall rates, but has sustainability challenges due to workforce shortages. Artificial intelligence (AI) as an independent reader (IR) in DR may provide a cost-effective solution with the potential to improve screening performance. Evidence for AI to generalise across different patient populations, screening programmes and equipment vendors, however, is still lacking. METHODS: This retrospective study simulated DR with AI as an IR, using data representative of real-world deployments (275,900 cases, 177,882 participants) from four mammography equipment vendors, seven screening sites, and two countries. Non-inferiority and superiority were assessed for relevant screening metrics. RESULTS: DR with AI, compared with human DR, showed at least non-inferior recall rate, cancer detection rate, sensitivity, specificity and positive predictive value (PPV) for each mammography vendor and site, and superior recall rate, specificity, and PPV for some. The simulation indicates that using AI would have increased arbitration rate (3.3% to 12.3%), but could have reduced human workload by 30.0% to 44.8%. CONCLUSIONS: AI has potential as an IR in the DR workflow across different screening programmes, mammography equipment and geographies, substantially reducing human reader workload while maintaining or improving standard of care. TRIAL REGISTRATION: ISRCTN18056078 (20/03/2019; retrospectively registered).

Artificial Intelligence as Supporting Reader in Breast Screening: A Novel Workflow to Preserve Quality and Reduce Workload.

OBJECTIVE: To evaluate the effectiveness of a new strategy for using artificial intelligence (AI) as supporting reader for the detection of breast cancer in mammography-based double reading screening practice. METHODS: Large-scale multi-site, multi-vendor data were used to retrospectively evaluate a new paradigm of AI-supported reading. Here, the AI served as the second reader only if it agrees with the recall/no-recall decision of the first human reader. Otherwise, a second human reader made an assessment followed by the standard clinical workflow. The data included 280 594 cases from 180 542 female participants screened for breast cancer at seven screening sites in two countries and using equipment from four hardware vendors. The statistical analysis included non-inferiority and superiority testing of cancer screening performance and evaluation of the reduction in workload, measured as arbitration rate and number of cases requiring second human reading. RESULTS: Artificial intelligence as a supporting reader was found to be superior or noninferior on all screening metrics compared with human double reading while reducing the number of cases requiring second human reading by up to 87% (245 395/280 594). Compared with AI as an independent reader, the number of cases referred to arbitration was reduced from 13% (35 199/280 594) to 2% (5056/280 594). CONCLUSION: The simulation indicates that the proposed workflow retains screening performance of human double reading while substantially reducing the workload. Further research should study the impact on the second human reader because they would only assess cases in which the AI prediction and first human reader disagree.

Minor class splicing shapes the zebrafish transcriptome during development.

Minor class or U12-type splicing is a highly conserved process required to remove a minute fraction of introns from human pre-mRNAs. Defects in this splicing pathway have recently been linked to human disease, including a severe developmental disorder encompassing brain and skeletal abnormalities known as Taybi-Linder syndrome or microcephalic osteodysplastic primordial dwarfism 1, and a hereditary intestinal polyposis condition, Peutz-Jeghers syndrome. Although a key mechanism for regulating gene expression, the impact of impaired U12-type splicing on the transcriptome is unknown. Here, we describe a unique zebrafish mutant, caliban (clbn), with arrested development of the digestive organs caused by an ethylnitrosourea-induced recessive lethal point mutation in the rnpc3 [RNA-binding region (RNP1, RRM) containing 3] gene. rnpc3 encodes the zebrafish ortholog of human RNPC3, also known as the U11/U12 di-snRNP 65-kDa protein, a unique component of the U12-type spliceosome. The biochemical impact of the mutation in clbn is the formation of aberrant U11- and U12-containing small nuclear ribonucleoproteins that impair the efficiency of U12-type splicing. Using RNA sequencing and microarrays, we show that multiple genes involved in various steps of mRNA processing, including transcription, splicing, and nuclear export are disrupted in clbn, either through intron retention or differential gene expression. Thus, clbn provides a useful and specific model of aberrant U12-type splicing in vivo. Analysis of its transcriptome reveals efficient mRNA processing as a critical process for the growth and proliferation of cells during vertebrate development.

Autophagy induction is a Tor- and Tp53-independent cell survival response in a zebrafish model of disrupted ribosome biogenesis.

Ribosome biogenesis underpins cell growth and division. Disruptions in ribosome biogenesis and translation initiation are deleterious to development and underlie a spectrum of diseases known collectively as ribosomopathies. Here, we describe a novel zebrafish mutant, titania (tti(s450)), which harbours a recessive lethal mutation in pwp2h, a gene encoding a protein component of the small subunit processome. The biochemical impacts of this lesion are decreased production of mature 18S rRNA molecules, activation of Tp53, and impaired ribosome biogenesis. In tti(s450), the growth of the endodermal organs, eyes, brain, and craniofacial structures is severely arrested and autophagy is up-regulated, allowing intestinal epithelial cells to evade cell death. Inhibiting autophagy in tti(s450) larvae markedly reduces their lifespan. Somewhat surprisingly, autophagy induction in tti(s450) larvae is independent of the state of the Tor pathway and proceeds unabated in Tp53-mutant larvae. These data demonstrate that autophagy is a survival mechanism invoked in response to ribosomal stress. This response may be of relevance to therapeutic strategies aimed at killing cancer cells by targeting ribosome biogenesis. In certain contexts, these treatments may promote autophagy and contribute to cancer cells evading cell death.

TGF-betaRII rescues development of small intestinal epithelial cells in Elf3-deficient mice.

BACKGROUND & AIMS: ELF3, a member of the ETS transcription factor family, has been shown to transactivate the transforming growth factor beta type II receptor (TGF-betaRII) promoter. Previously we showed that Elf3-null mice have a defect in the small intestine caused by a failure of small intestinal epithelial cells to differentiate and that these cells produced significantly lower levels of Tgf-betaRII. To prove that the defect observed in Elf3-null mice resulted from the lack of Elf3-dependent activation of Tgf-betaRII expression, we performed a genetic rescue. METHODS: We generated transgenic mice that express human TGF-betaRII specifically in the intestinal epithelium under the control of the mouse A33 antigen promoter. Mice expressing the A33-TGF-betaRII transgene were mated with Elf3(+/-) mice, and double heterozygous offspring harboring both the transgene and one mutant Elf3 allele were intercrossed. RESULTS: The resultant A33-TGF-betaRII transgenic Elf3(-/-) pups displayed normal small intestinal morphology, while the characteristic abnormality was retained in all Elf3(-/-) mice that did not express the transgene. This phenotypic rescue shows for the first time in vivo that a single gene, Elf3, is the critical upstream regulator of Tgf-betaRII in mouse small intestinal epithelium. CONCLUSIONS: This has important implications for our understanding of tissue-specific gene regulation and further strengthens the potential clinical connection between ELF3 and colorectal cancer involving transforming growth factor beta insensitivity.

Inactivation of the transcription factor Elf3 in mice results in dysmorphogenesis and altered differentiation of intestinal epithelium.

BACKGROUND & AIMS: The mammalian small intestine is lined by a highly specialized epithelium that functions in the digestion and absorption of nutrients. The molecular mechanisms that direct intestinal epithelial cell morphogenesis and terminal differentiation are poorly understood. We have previously identified Elf3 (E74-like factor-3) as a member of the ETS transcription factor family strongly expressed in small intestinal epithelium. The aim of this study is to investigate the biological roles of Elf3 in vivo. METHODS: Mice with a null mutation of Elf3 were generated through targeted gene disruption. Characterization of intestinal development was performed by histologic and immunohistochemical techniques. RESULTS: Targeted disruption of Elf3 resulted in fetal lethality of about 30% at around embryonic day 11.5. Seventy percent of the Elf3-deficent progeny were born and displayed severe alterations of tissue architecture in the small intestine, manifested by poor villus formation and abnormal morphogenesis and terminal differentiation of absorptive enterocytes and mucus-secreting goblet cells. Crypt cell proliferation, however, appeared intact in Elf3-deficient mice.Elf3-deficient enterocytes express markedly reduced levels of the transforming growth factor beta type II receptor (TGF-beta RII), an inducer of intestinal epithelial differentiation. CONCLUSIONS: Elf3 is an important regulator of morphogenesis and terminal differentiation of epithelial cell lineages in the small intestine.