Are better AI algorithms for breast cancer detection also better at predicting risk? A paired case-control study.

BACKGROUND: There is increasing evidence that artificial intelligence (AI) breast cancer risk evaluation tools using digital mammograms are highly informative for 1-6 years following a negative screening examination. We hypothesized that algorithms that have previously been shown to work well for cancer detection will also work well for risk assessment and that performance of algorithms for detection and risk assessment is correlated. METHODS: To evaluate our hypothesis, we designed a case-control study using paired mammograms at diagnosis and at the previous screening visit. The study included n = 3386 women from the OPTIMAM registry, that includes mammograms from women diagnosed with breast cancer in the English breast screening program 2010-2019. Cases were diagnosed with invasive breast cancer or ductal carcinoma in situ at screening and were selected if they had a mammogram available at the screening examination that led to detection, and a paired mammogram at their previous screening visit 3y prior to detection when no cancer was detected. Controls without cancer were matched 1:1 to cases based on age (year), screening site, and mammography machine type. Risk assessment was conducted using a deep-learning model designed for breast cancer risk assessment (Mirai), and three open-source deep-learning algorithms designed for breast cancer detection. Discrimination was assessed using a matched area under the curve (AUC) statistic. RESULTS: Overall performance using the paired mammograms followed the same order by algorithm for risk assessment (AUC range 0.59-0.67) and detection (AUC 0.81-0.89), with Mirai performing best for both. There was also a correlation in performance for risk and detection within algorithms by cancer size, with much greater accuracy for large cancers (30 mm+, detection AUC: 0.88-0.92; risk AUC: 0.64-0.74) than smaller cancers (0 to < 10 mm, detection AUC: 0.73-0.86, risk AUC: 0.54-0.64). Mirai was relatively strong for risk assessment of smaller cancers (0 to < 10 mm, risk, Mirai AUC: 0.64 (95% CI 0.57 to 0.70); other algorithms AUC 0.54-0.56). CONCLUSIONS: Improvements in risk assessment could stem from enhancing cancer detection capabilities of smaller cancers. Other state-of-the-art AI detection algorithms with high performance for smaller cancers might achieve relatively high performance for risk assessment.

Predator-induced changes of male and female mating preferences: innate and learned components.

While many mating preferences have a genetic basis, the question remains as to whether and how learning/experience can modify individual mate choice decisions. We used wild-caught (predator-experienced) and F1 laboratory-reared (predator-naïve) invasive Western mosquitofish Gambusia affinis from China to test whether mating preferences (assessed in a first mate choice test) would change under immediate predation threat. The same individuals were tested in a second mate choice test during which 1 of 3 types of animated predators was presented: 1) a co-occurring predator, 2) a co-evolved but not currently co-occurring predator, and 3) a non-piscivorous species as control. We compared preference scores derived from both mate choice tests to separate innate from experiential effects of predation. We also asked whether predator-induced changes in mating preferences would differ between sexes or depend on the choosing individual's personality type and/or body size. Wild-caught fish altered their mate choice decisions most when exposed to the co-occurring predator whereas laboratory-reared individuals responded most to the co-evolved predator, suggesting that both innate mechanisms and learning effects are involved. This behavior likely reduces individuals' risk of falling victim to predation by temporarily moving away from high-quality (i.e., conspicuous) mating partners. Accordingly, effects were stronger in bolder than shyer, large- compared with small-bodied, and female compared with male focal individuals, likely because those phenotypes face an increased predation risk overall. Our study adds to the growing body of literature appreciating the complexity of the mate choice process, where an array of intrinsic and extrinsic factors interacts during decision-making.