similar with respect to the preference rates for life-sustaining treatments compared with palliative care (46.9% vs 34.4% in the 60% survival group and 50.0% vs 40.6% in the 30% survival group; odds ratio [OR], 0.90; 95% CI, 0.31-2.63). A few patients were not able to formulate a preference (6 patients (18.8%) in the 60% survival group and 3 patients (9.4%) in the 30% survival group; OR, 0.423; 95% CI, 0.08-2.10). An analysis of the patients who formulated a preference showed that an attitude that mere survival is at least as important as quality of life was associated with a preference for life-sustaining treatments (OR, 10.28; 95% CI, 2.94-35.90). Increasing maternal age (OR, 0.77; 95% CI, 0.61-0.98) and childlessness (OR, 0.12; 95% CI 0.01-0.98) were associated with a preference for palliative care. Most patients would decide together with their partners (63 of 64 [98.4%]) and preferred to be empowered by their physicians in the decision-making process (48 of 64 [75%]). Discussion | In this study, it appeared that treatment preferences originated from individual characteristics and values rather than from reasoning about numerical outcome estimates. However, generalizability is limited and the results should be interpreted in light of the methods used. Patients made a one-time decision without personal feedback and patients actually affected might indicate different preferences. More studies are needed to help to improve our understanding of the information that parents facing extremely preterm birth want and need.

Objective: Axillary lymph node (ALN) metastasis status is important in guiding treatment in breast cancer. The aims were to assess how deep convolutional neural network (CNN) performed compared with radiomics analysis in predicting ALN metastasis using breast ultrasound, and to investigate the value of both intratumoral and peritumoral regions in ALN metastasis prediction. Methods:We retrospectively enrolled 479 breast cancer patients with 2,395 breast ultrasound images. Based on the intratumoral, peritumoral, and combined intra-and peritumoral regions, three CNNs were built using DenseNet, and three radiomics models were built using random forest, respectively. By combining the molecular subtype, another three CNNs and three radiomics models were built. All models were built on training cohort (343 patients 1,715 images) and evaluated on testing cohort (136 patients 680 images) with ROC analysis. Another prospective cohort of 16 patients was enrolled to further test the models.Results: AUCs of image-only CNNs in both training/testing cohorts were 0.957/0.912 for combined region, 0.944/0.775 for peritumoral region, and 0.937/0.748 for intratumoral region, which were numerically higher than their corresponding radiomics models with AUCs of 0.940/0.886, 0.920/0.724, and 0.913/0.693. The overall performance of image-molecular CNNs in terms of AUCs on training/testing cohorts slightly increased to 0.962/0.933, 0.951/0.813, and 0.931/0.794, respectively. AUCs of both CNNs and radiomics models built on combined region were significantly better than those on either intratumoral or peritumoral region on the testing cohort (p < 0.05). In the prospective study, the CNN model built on combined region achieved the highest AUC of 0.95 among all image-only models. Sun et al. Ultrasound-CNN Predicted Breast Cancer MetastasisConclusions: CNNs showed numerically better overall performance compared with radiomics models in predicting ALN metastasis in breast cancer. For both CNNs and radiomics models, combining intratumoral, and peritumoral regions achieved significantly better performance.

No abstract

OBJECTIVES: To determine the diagnostic and clinical utility of trio-rapid genome sequencing in critically ill infants. DESIGN: In this prospective study, samples from critically ill infants were analyzed using both proband-only clinical exome sequencing and trio-rapid genome sequencing (proband and biological parents). The study occurred between April 2019 and December 2019. SETTING: Thirteen member hospitals of the China Neonatal Genomes Project spanning 10 provinces were involved. PARTICIPANTS: Critically ill infants (n = 202), from birth up until 13 months of life were enrolled based on eligibility criteria (e.g., CNS anomaly, complex congenital heart disease, evidence of metabolic disease, recurrent severe infection, suspected immune deficiency, and multiple malformations). INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Of the 202 participants, neuromuscular (45%), respiratory (22%), and immunologic/infectious (18%) were the most commonly observed phenotypes. The diagnostic yield of trio-rapid genome sequencing was higher than that of proband-only clinical exome sequencing (36.6% [95% CI, 30.1–43.7%] vs 20.3% [95% CI, 15.1–26.6%], respectively; p = 0.0004), and the average turnaround time for trio-rapid genome sequencing (median: 7 d) was faster than that of proband-only clinical exome sequencing (median: 20 d) (p < 2.2 × 10–16). The metagenomic analysis identified pathogenic or likely pathogenic microbes in six infants with symptoms of sepsis, and these results guided the antibiotic treatment strategy. Sixteen infants (21.6%) experienced a change in clinical management following trio-rapid genome sequencing diagnosis, and 24 infants (32.4%) were referred to a new subspecialist. CONCLUSIONS: Trio-rapid genome sequencing provided higher diagnostic yield in a shorter period of time in this cohort of critically ill infants compared with proband-only clinical exome sequencing. Precise and fast molecular diagnosis can alter medical management and positively impact patient outcomes.