A generalised approach for high-throughput instance segmentation of stomata in microscope images

BACKGROUND

Stomata analysis using microscope imagery provides important insight into plant physiology, health and the surrounding environmental conditions. Plant scientists are now able to conduct automated high-throughput analysis of stomata in microscope data, however, existing detection methods are sensitive to the appearance of stomata in the training images, thereby limiting general applicability. In addition, existing methods only generate bounding-boxes around detected stomata, which require users to implement additional image processing steps to study stomata morphology. In this paper, we develop a fully automated, robust stomata detection algorithm which can also identify individual stomata boundaries regardless of the plant species, sample collection method, imaging technique and magnification level.

RESULTS

The proposed solution consists of three stages. First, the input image is pre-processed to remove any colour space biases occurring from different sample collection and imaging techniques. Then, a Mask R-CNN is applied to estimate individual stomata boundaries. The feature pyramid network embedded in the Mask R-CNN is utilised to identify stomata at different scales. Finally, a statistical filter is implemented at the Mask R-CNN output to reduce the number of false positive generated by the network. The algorithm was tested using 16 datasets from 12 sources, containing over 60,000 stomata. For the first time in this domain, the proposed solution was tested against 7 microscope datasets never seen by the algorithm to show the generalisability of the solution. Results indicated that the proposed approach can detect stomata with a precision, recall, and F-score of 95.10%, 83.34%, and 88.61%, respectively. A separate test conducted by comparing estimated stomata boundary values with manually measured data showed that the proposed method has an IoU score of 0.70; a 7% improvement over the bounding-box approach.

CONCLUSIONS

The proposed method shows robust performance across multiple microscope image datasets of different quality and scale. This generalised stomata detection algorithm allows plant scientists to conduct stomata analysis whilst eliminating the need to re-label and re-train for each new dataset. The open-source code shared with this project can be directly deployed in Google Colab or any other Tensorflow environment.

Accelerating Automated Stomata Analysis Through Simplified Sample Collection and Imaging Techniques

Digital image processing is commonly used in plant health and growth analysis, aiming to improve research efficiency and repeatability. One focus is analysing the morphology of stomata, with the aim to better understand the regulation of gas exchange, its link to photosynthesis and water use and how they are influenced by climatic conditions. Despite the key role played by these cells, their microscopic analysis is largely manual, requiring intricate sample collection, laborious microscope application and the manual operation of a graphical user interface to identify and measure stomata. This research proposes a simple, end-to-end solution which enables automatic analysis of stomata by introducing key changes to imaging techniques, stomata detection as well as stomatal pore area calculation. An optimal procedure was developed for sample collection and imaging by investigating the suitability of using an automatic microscope slide scanner to image nail polish imprints. The use of the slide scanner allows the rapid collection of high-quality images from entire samples with minimal manual effort. A convolutional neural network was used to automatically detect stomata in the input image, achieving average precision, recall and F-score values of 0.79, 0.85, and 0.82 across four plant species. A novel binary segmentation and stomatal cross section analysis method is developed to estimate the pore boundary and calculate the associated area. The pore estimation algorithm correctly identifies stomata pores 73.72% of the time. Ultimately, this research presents a fast and simplified method of stomatal assay generation requiring minimal human intervention, enhancing the speed of acquiring plant health information.

Stratification of fertility potential according to cervical mucus symptoms: achieving pregnancy in fertile and infertile couples

Women wishing to conceive are largely unaware of fertility symptoms at the time of ovulation. This study investigated the effectiveness of fertility-awareness in achieving pregnancy, particularly fertile mucus pattern, in the context of infertility. The 384 eligible participants were taken from consecutive women desiring pregnancy who attended 17 Australian Billings Ovulation Method^® clinics from 1999-2003. This cohort included couples with infertility ≥12 months (51%) and female age >35 years (28%). Under fertility-awareness instruction, pregnancy was achieved by 240 couples (62.5%) after maximum follow-up of two years. Mucus symptom observations effectively stratified 'low pregnancy-potential' (35.2%) and 'high pregnancy-potential' groups. Pregnancy rates were ∼30% higher in the latter group (44.4% versus 72.3%) in addition to consistent effects observed on pregnancy achievements within subgroups defined by prognostic factors such as duration of infertility (p = 0.001) and increasing female age (p = 0.04). Fertile symptoms were also associated with significantly shorter time to conception (4.2 versus 6.4 months) in a survival analysis (p = 0.003). Billings Ovulation Method^® observations strongly predicted successful conception. This has the capacity to provide a rapid, reliable and cost-effective approach to stratifying fertility potential, including directing timely and targeted investigations/management, and is accessible for women who may be remote from primary or specialist care.

Microscope image based fully automated stomata detection and pore measurement method for grapevines

BACKGROUND

Stomatal behavior in grapevines has been identified as a good indicator of the water stress level and overall health of the plant. Microscope images are often used to analyze stomatal behavior in plants. However, most of the current approaches involve manual measurement of stomatal features. The main aim of this research is to develop a fully automated stomata detection and pore measurement method for grapevines, taking microscope images as the input. The proposed approach, which employs machine learning and image processing techniques, can outperform available manual and semi-automatic methods used to identify and estimate stomatal morphological features.

RESULTS

First, a cascade object detection learning algorithm is developed to correctly identify multiple stomata in a large microscopic image. Once the regions of interest which contain stomata are identified and extracted, a combination of image processing techniques are applied to estimate the pore dimensions of the stomata. The stomata detection approach was compared with an existing fully automated template matching technique and a semi-automatic maximum stable extremal regions approach, with the proposed method clearly surpassing the performance of the existing techniques with a precision of 91.68% and an F1-score of 0.85. Next, the morphological features of the detected stomata were measured. Contrary to existing approaches, the proposed image segmentation and skeletonization method allows us to estimate the pore dimensions even in cases where the stomatal pore boundary is only partially visible in the microscope image. A test conducted using 1267 images of stomata showed that the segmentation and skeletonization approach was able to correctly identify the stoma opening 86.27% of the time. Further comparisons made with manually traced stoma openings indicated that the proposed method is able to estimate stomata morphological features with accuracies of 89.03% for area, 94.06% for major axis length, 93.31% for minor axis length and 99.43% for eccentricity.

CONCLUSIONS

The proposed fully automated solution for stomata detection and measurement is able to produce results far superior to existing automatic and semi-automatic methods. This method not only produces a low number of false positives in the stomata detection stage, it can also accurately estimate the pore dimensions of partially incomplete stomata images. In addition, it can process thousands of stomata in minutes, eliminating the need for researchers to manually measure stomata, thereby accelerating the process of analysing plant health.

Age/sex/location: uncovering the social cues in the development of online relationships

Past research on online relationships has predominantly been concerned with how the quality of online relationships compares with offline relationships. This research has been more concerned with the medium itself than with the meanings that users construct around their interpersonal interactions within this medium. The current paper seeks to redress this imbalance by exploring the ways that available social cues are used to shape the meanings of online relationships. Sixty Internet users, ranging in age from 19-51 years, were interviewed about their online relationships. It was found that ideals that are important in traditional relationships, such as trust, honesty, and commitment are just as important in online relationships; however, the cues that signify these ideals vary.

Systemic and renovascular hypertension

To ascertain the contribution of systemic hypertension in the progression of renal failure, we have studied the effects of pharmacological treatment of hypertension in rats with the remnant kidney model of renal insufficiency, streptozotocin diabetes, or nephrotoxic serum nephritis. Treatment with the angiotensin converting enzyme (ACE) inhibitor enalapril lowered systemic blood pressure in the remnant kidney and diabetic animals, but did not lower blood pressure in rats with nephrotoxic serum nephritis. Proteinuria was reduced in all three models, and creatinine clearance improved in the remnant kidney and diabetic animals, when compared with untreated controls. In the remnant kidney and diabetic models systemic blood pressure was lowered to a similar degree by treatments with a calcium blocker, with no improvement in either proteinuria, or glomerular filtration rate. Further studies of the long-term effects of enalapril have been undertaken in rats with the two kidney one clip model of hypertension. Rats treated with enalapril had a lower blood pressure and improved survival over one year of treatment, compared with untreated rats. After 1 year of treatment however the clipped kidney was small and fibrotic, and non functional. Following withdrawal of enalapril therapy there was no functional improvement of the clipped kidney. The possibility that ACE inhibitors have a specific intra-renal effect reducing the rate of progression of renal disease now needs confirmation in human studies. In renovascular hypertension however, intra-renal changes induced by ACE inhibitors may cause irreversible renal damage.

Preservation of renal structure and function in the rat remnant kidney model of chronic renal failure by enalapril treatment

The remnant kidney model of chronic renal failure was established in rats subject to subtotal (1 7/8) nephrectomy and the evolution of renal injury studied over a period of 6 wk. One wk after subtotal nephrectomy, rats had a mean conscious systolic blood pressure of 158 +/- 5 mm Hg and serum creatinine of 128 +/- 9 mumol/l. Both systolic blood pressure and serum creatinine rose over the next 5 wk in concert with progressive glomerulosclerosis and proteinuria. Enalapril, an angiotensin converting enzyme inhibitor, was administered (5 mg/kg/day) to rats (n = 11) from 1 wk after subtotal nephrectomy. Enalapril lowered systolic blood pressure over the treatment period. Systolic blood pressure was 122 +/- 5 mm Hg compared with 176 +/- 7 mm Hg in untreated rats (p less than 0.001) at 6 wk. Serum creatinine 6 wk after subtotal nephrectomy was 110 +/- 9 mumol/l with enalapril treatment, compared with 159 +/- 21 mumol/l (p less than 0.025) in control animals. Enalapril treated rats had lower urinary protein excretion than controls (15 +/- 3 mg/24 hr vs 85 +/- 22 mg/24 hr, p less than 0.0001) at 6 weeks. Glomerulosclerosis, assessed by blinded histological score, was also reduced in the enalapril treated group (1.79 +/- 0.08 vs 2.36 +/- 0.16, p less than 0.01). Enalapril treatment was associated with a reduction in filtration fraction (51Cr-EDTA/125I-hippurate clearance). At 6 wk, filtration fraction was 0.30 +/- 0.03 in enalapril treated and 0.48 +/- 0.03 in control rats (p less than 0.001). Enalapril treatment in the subtotal nephrectomy model of renal failure preserved renal structure and function.(ABSTRACT TRUNCATED AT 250 WORDS)

Preservation of renal function in the rat remnant kidney model of chronic renal failure by blood pressure reduction

The remnant kidney model of chronic renal failure was established in female Sprague-Dawley rats subjected to surgical removal of the right kidney and segmental infarction of seven-eighths of the left kidney. Plasma creatinine (mumol/l) rose from 65 (s.e.m. = 16, n = 18) to 153 (s.e.m. = 27, t-test, P less than 0.001, d.f. = 17) over 6 weeks. Histological glomerulosclerotic lesions were present from 2 weeks and prominent by 6 weeks post-surgery. Rats were treated with enalapril (5 mg/kg per day, n = 11) or felodipine (30 mg/kg per day, n = 13) from 1 week post-surgery, and their course compared to untreated rats (n = 18). Blood pressure (mmHg) was lowered by both treatments. Six weeks post-nephrectomy, systolic blood pressure in the untreated group was 176 (s.e.m. = 7, n = 18), enalapril group 122 (s.e.m. = 5, t-test, P less than 0.001, d.f. = 27), and felodipine group 128 (s.e.m. = 3, t-test, P less than 0.001, d.f. = 29). Plasma creatinine (mumol/l) was lower in the enalapril group (110, s.e.m. = 8, t-test, P less than 0.05, d.f. = 27) but not the felodipine group (173, s.e.m. = 19, t-test, n.s.) 6 weeks after subtotal nephrectomy compared to the untreated group (153, s.e.m. = 27). Glomerulosclerosis (blinded histological score) was reduced with enalapril treatment (1.93, s.e.m. = 0.03, t-test, P less than 0.05, d.f. = 27) but not felodipine treatment (2.15, s.e.m. = 0.04, c.f. untreated 2.36, s.e.m. = 0.12, t-test, n.s.). Urinary protein excretion (mg/24 h) was 84 (s.e.m. = 22, n = 13) in untreated rats, 15 (s.e.m. = 3, t-test, P less than 0.001, d.f. = 22) in enalapril-treated rats and 221 (s.e.m. = 35, n = 10) with felodipine treatment. Functional and structural damage in the rat remnant kidney model of chronic renal failure was ameliorated by treatment with enalapril but not by treatment with felodipine.