Deep learning-based classification of the mouse estrous cycle stages

Abstract

There is a rapidly growing demand for female animals in preclinical animal, and thus it is necessary to determine animals' estrous cycle stages from vaginal smear cytology. However, the determination of estrous stages requires extensive training, takes a long time, and is costly; moreover, the results obtained by human examiners may not be consistent. Here, we report a machine learning model trained with 2,096 microscopic images that we named the "Stage Estimator of estrous Cycle of RodEnt using an Image-recognition Technique (SECREIT)." With the test dataset (736 images), SECREIT achieved area under the receiver-operating-characteristic curve of 0.962 or more for each estrous stage. A test using 100 images showed that SECREIT provided correct classification that was similar to that provided by two human examiners (SECREIT: 91%, Human 1: 91%, Human 2: 79%) in 11 s. The SECREIT can be a first step toward accelerating the research using female rodents. Knowledge of the precise stages of the estrous cycle is very important for interpretations of female animals' data. Compared to men, women have a high lifetime incidence of several mental illnesses, including depression, posttraumatic stress disorders, generalized anxiety, and eating disorders 1 , but preclinical animal investigations for these illnesses have use mainly males in part because of the volatility of female animals' experimental data that cannot be separated from their estrous cycle. Several research groups have stated that it is difficult to apply the findings from males' preclinical results to women's medicine 2. Indeed, the estrous cycle affects the expression of genes 3 , proteins 4,5 , electrophysiological properties 6,7 , behaviors 8,9 , and drug effects 10. In 2015, the U.S. National Institutes of Health (NIH) announced that when conducting NIH-funded research, researchers should study both sexes 11. The number of preclinical studies using female animals is thus gradually increasing. The estrous cycle in rodents is generally divided into three or four stages 12 , and the cycle is 4-5 days: Diestrus (D) → proestrus (P) → estrus (E) → (metestrus) →. In many studies of rodents, the estrous cycle stage of each animal has been determined by vaginal cytology. Each stage is decided based on the type, number, shape, size, and proportion of cells in a vaginal smear 12-17 (Fig. 1a-c). Briefly, stage D was identified by the presence of leukocytes and nucleated cells with or without a few cornified cells. Stage P was identified by the presence of nucleated epithelial cells and cornified cells without leukocytes. Stage E was identified by the presence of nucleated epithelial cells without leukocytes or cornified cells. Vaginal cytology is also used to decide whether an ovariectomy in a rodent is successful 18,19 (Fig. 1d). However, determining the estrous stage of a rodent by using vaginal cytology evaluated by a human examiner has some problems: (1) a long training period is required in order to become skillful; (2) it takes a long time to determine the estrou...

Effect of Estrous Cycle on Behavior of Females in Rodent Tests of Anxiety

“…Several attempts have been made to develop rapid, objective methods, particularly for use in mice. These include using non-stained smear material ( 27 , 40 ), modifications to staining methods ( 41 ) and application of deep learning technology for classification of smears ( 42 ). Alternatives to vaginal cytology have been proposed based on gross examination of the vaginal opening ( 43 ), changes in skin temperature due to activation of brown adipose tissue ( 44 ) and variations in vaginal wall impedence ( 45 ).…”

Automated classification of estrous stage in rodents using deep learning

“…Although previous studies have used computational methods to analyze vaginal cytology 12,30 , the input datasets for these networks have historically been restricted to a single stain. To further enhance generalizability, the training and validation image sets for EstrousNet include samples stained with H&E, Shorr, Giemsa, cresyl violet, and crystal violet stains, at magnifications of 10x and 20x ( Fig.…”

“…Other CNNs trained on 3 stages from a single dataset therefore exhibit higher validation accuracy in some stages 12,30 . Additionally, the fourth and most transient stage of the estrous cycle, metestrus, yields the lowest test accuracy, as is consistent with previously developed machine learning approaches 12 . Since the presence of both cornified and nucleated epithelial cells in metestrus causes confusion with proestrus, more data will be useful for training CNNs to differentiate between these two stages.…”

“…In female rats and mice, the analogous estrous cycle lasts 4-5 days 10 , but exhibits steroid hormone fluctuations similar to the 28 day human menstrual cycle. The estrous cycle was first described over a century ago 11 , yet the criteria for tracking this cycle remain subjective and variable between experimenters 12 . Determining the stage of estrous is critical to evaluating the state of the hypothalamic-pituitary-ovarian axis, which has implications in a myriad of factors including gene expression 13,14 , neuronal structure and connectivity 3,15 , and pharmacological efficacy 16 .…”

“…2) it lacks generalizability; even expert classifiers may have trouble generalizing across rodent species, stains, and subjects. 3) it is inconsistent between labs, as classification can vary widely between human examiners 12 . Here, we address these challenges using a novel deep learning algorithm that can generate estrous stage classifications on the order of seconds.…”