Deep learning-based high-throughput detection of in vitro germination to assess pollen viability from microscopic images.

In vitro pollen germination is considered the most efficient method to assess pollen viability. The pollen germination frequency and pollen tube length, which are key indicators of pollen viability, should be accurately measured during in vitro culture. In this study, a Mask R-CNN model trained using microscopic images of tree peony (Paeonia suffruticosa) pollen has been proposed to rapidly detect the pollen germination rate and pollen tube length. To reduce the workload during image acquisition, images of synthesized crossed pollen tubes were added to the training dataset, significantly improving the model accuracy in recognizing crossed pollen tubes. At an Intersection over Union threshold of 50%, a mean average precision of 0.949 was achieved. The performance of the model was verified using 120 testing images. The R2 value of the linear regression model using detected pollen germination frequency against the ground truth was 0.909 and that using average pollen tube length was 0.958. Further, the model was successfully applied to two other plant species, indicating a good generalizability and potential to be applied widely.

Clinical, epidemiological and microbiological characteristics of relapse and re-infection in Clostridium difficile infection.

Recurrent diarrhea is a common complication of Clostridium difficile infection (CDI). Recurrent CDI (r-CDI) may be produced by the persistence of spores (relapse) or by the acquisition of a new strain (reinfection). In this study, we analyze epidemiological, clinical, microbiological and laboratory data from patients with r-CDI, relapse, and reinfection-CDI over 5 years and compared with a control group (non r-CDI). Among 60 patients with r-CDI, 36 patients had stool samples collected from two or more episodes, which were molecularly analyzed. Based on ribotyping, 63.9% of the samples were relapse, and 36.1% reinfection. In a multivariable logistic regression analysis, previous antibiotic exposure was found to be a risk factor for r-CDI (OR: 2.23; 95% CI: 1.0-4.9; p = 0.04). Patients with relapse had previous antibiotic exposure more frequently than did patients with reinfection (p = 0.03), and patients with reinfection suffered more frequently from chronic liver disease (p = 0.02) than did relapse patients. Relapse patients compared with the control group had a higher percentage of previous antibiotic exposure, although the difference was statistically no significant (73.9% vs. 91.3 p = 0.06). No significant differences for the selected variables were observed between the reinfection and control groups, although we observed a higher percentage of patients with chronic liver disease (30.8% vs 13.3%; p = 0.08). All isolates were sensitive to metronidazole and vancomycin. No significant differences in antibiotic susceptibility were found between the different groups. Sporulation and germination frequency of r-CDI were higher than non r-CDI (p = 0.02 and p < 0.01, respectively). Nevertheless, there were statistically not significant differences between the relapse and reinfection groups. Both frequencies were compared between the first and second episode of CDI for the relapse and reinfection groups, but differences were not observed to be statistically significant. In conclusion, our study showed that the recurrence of CDI was associated with antibiotic use and sporulation/germination frequency, regardless of relapse or reinfection. The use of antibiotics would produce a dysbiosis and favor the persistence of the C. difficile spores and relapse. A possible alteration of the intestinal microbiota and the bile salts produced by chronic liver disease could favor reinfection.