Development of Growth Estimation Algorithms for Hydroponic Bell Peppers Using Recurrent Neural Networks

As smart farms are applied to agricultural fields, the use of big data is becoming important. In order to efficiently manage smart farms, relationships between crop growth and environmental conditions are required to be analyzed. From this perspective, various artificial intelligence algorithms can be used as useful tools to quantify this relationship. The objective of this study was to develop and validate an algorithm that can interpret the crop growth rate response to environmental factors based on a recurrent neural network (RNN), and to evaluate the algorithm accuracy compared to the process-based model (PBM). The algorithms were trained with data from three growth periods. The developed methods were used to measure the crop growth rate. The algorithm consisted of eight environmental variables days after transplanting and two crop growth characteristics as input variables producing weekly crop growth rates as output. The RNN-based crop growth rate estimation algorithm was validated using data collected from a commercial greenhouse. The CropGro-bell pepper model was applied to compare and evaluate the accuracy of the developed algorithm. The training accuracies varied from 0.75 to 0.81 in all growth periods. From the validation result, it was confirmed that the accuracy was reliable in the commercial greenhouse. The accuracy of the developed algorithm was higher than that of the PBM. The developed algorithm can contribute to crop growth estimation with a limited number of data.

L. G. Ramensky and allometry of plants (history and current state of the problem)

A historical review of the new direction of botanical science, namely, «plant allometry» was made. In this review, the attention is focused only on establishing the dependence of the aboveground mass of plants on their size. The first foreign experiments on the indirect determination of the aboveground mass of herbs, semishrub, and low shrub date back to the 1930s. Now the determination of the aboveground mass of these plants based on the re-sults of measuring their habitus has become widespread. In addition to height and cover, the volume of plants began to be used as predictors of aboveground biomass. Since the second half of the last century, power (expo-nential) equations have been used to estimate the aboveground mass of plants in hayfields and pastures from the results of measuring their habitus. By now, the "allometry of plants" has gone far beyond the limits, which reflect only the regularities connecting the sizes of plants with each other, as well as their size and productivity of plants. L. G. Ramensky began to conduct his experiments on the indirect determination of the aboveground mass of plants in hayfields and pastures much earlier than foreign researchers did. In 1915, he proposed the notion of "projective weight". This is the mass of plants per unit area of its cover. At first L. G. Ramensky assumed to calculate this value proceeding from the projective cover of plants. However, soon he became convinced that such a calculation gives a very unstable result. In 1938, L. G. Ramensky proposed to take into account the height of plants to determine the projective weight. He also proposed multiple linear regression equations to predict projective weight depending on the height of vegetative shoots and the ratio of the number of flower shoots to the cover of the plant. In the early 1950s, L. G. Ramensky summarized the accumulated material regarding the de-termination of the weight of the aboveground mass of individual plants by the parameters of their habitus. How-ever, this generalization was published only in 1966, 13 years after the death of L. G. Ramensky. Together with his colleagues, he calculated a special coefficient for many plants, reflecting the degree of dependence of the projective weight on the morphological, anatomical features of the species, as well as the conditions of their growth. Estimation of the aboveground mass of herbaceous plants, semishrub and low shrub based on the results of measuring their habitus that was started by Ramensky at the beginning of the last century, has outgrown the utilitarian need to determine feed reserves in hayfields and pastures. This assessment entered the framework of an independent direction of botanical science, which is currently called «plant allometry».

A 3D analysis of growth trajectory and integration during early human prenatal facial growth

Significant shape changes in the human facial skeleton occur in the early prenatal period, and understanding this process is critical for studying a myriad of congenital facial anomalies. However, quantifying and visualizing human fetal facial growth has been challenging. Here, we applied quantitative geometric morphometrics (GM) to high-resolution magnetic resonance images of human embryo and fetuses, to comprehensively analyze facial growth. We utilized non-linear growth estimation and GM methods to assess integrated epigenetic growth between masticatory muscles and associated bones. Our results show that the growth trajectory of the human face in the early prenatal period follows a curved line with three flexion points. Significant antero-posterior development occurs early, resulting in a shift from a mandibular prognathic to relatively orthognathic appearance, followed by expansion in the lateral direction. Furthermore, during this time, the development of the zygoma and the mandibular ramus is closely integrated with the masseter muscle.

Oldest Indo-Pacific Lionfish (Pterois volitans/P. miles) Recorded From the Northwestern Gulf of Mexico

Indo-Pacific lionfish (Pterois volitans/miles) were first detected off the coast of Florida in the 1980s, with aquaria release being the most likely mechanism for introduction. Since then, lionfish have proliferated through the Western Atlantic Ocean, Caribbean Sea, and Gulf of Mexico (GOM). Here, we report the oldest lionfish aged on record in the Western Atlantic, removed from Flower Garden Banks National Marine Sanctuary (FGBNMS) in the GOM. In August 2018, a research expedition removed 745 lionfish from FGBNMS, of which a subset were retained for age and growth estimation. The oldest lionfish was a 10 y old male, with total length 375 mm and weight 805 g. The back-calculated birth date (2008) preceded the first observation of lionfish at FGBNMS by 3 years (2011). It is not well understood if lionfish are having negative impacts at FGBNMS, but this report signifies the importance of continued monitoring and removal efforts of this protected area.

TMOD-09. GLIOBLASTOMA BIOPHYSICAL GROWTH ESTIMATION USING DEEP LEARNING-BASED REGRESSION

BACKGROUND Glioblastomas are arguably the most aggressive, infiltrative, and heterogeneous adult brain tumor. Biophysical modeling of glioblastoma growth has shown its predictive value towards clinical endpoints, enabling more informed decision-making. However, the mathematically rigorous formulations of biophysical modeling come with a large computational footprint, hindering their application to clinical studies. METHODS We present a deep learning (DL)-based logistical regression model, to estimate in seconds glioblastoma biophysical growth, defined through three tumor-specific parameters: 1) diffusion coefficient of white matter (Dw), which describes how easily the tumor can infiltrate through the white matter, 2) mass-effect parameter (Mp), which defines the average tumor expansion, and 3) estimated time (T) in number of days that the tumor has been growing. Pre-operative multi-parametric MRI (mpMRI) structural scans (T1, T1-Gd, T1, T2-FLAIR) from 135 subjects of the TCGA-GBM imaging collection are used to quantitatively evaluate our approach. We consider the mpMRI intensities within the region defined by the abnormal T2-FLAIR signal envelope, for training three DL models for the three tumor-specific parameters. Each of our DL models consist of two sets of convolution layers followed by a single max-pooling layer, with a normalized root mean squared error as the minimization metric and evaluated using 10-fold cross validation. We train and validate the DL-based predictions against parameters derived from biophysical inversion models. RESULTS Pearson correlation coefficients between our DL-based estimations and the biophysical parameters were equal to 0.85 for Dw, 0.90 for Mp, and 0.94 for T. CONCLUSION This study unlocks the power of tumor-specific parameters from biophysical tumor growth estimation, paving the way towards their utilization in more clinical studies, while opening the door for leveraging advanced radiomic descriptors in future studies, as well as allowing for significantly faster parameter reconstruction compared to biophysical growth modeling approaches. *denotes equal senior authorship.