StressNet: a spatial-spectral-temporal deformable attention-based framework for water stress classification in maize.

In recent years, monitoring the health of crops has been greatly aided by deploying highthroughput crop monitoring techniques that integrate remotely captured imagery and deep learning techniques. Most methods rely mainly on the visible spectrum for analyzing the abiotic stress, such as water deficiency in crops. In this study, we carry out experiments on maize crop in a controlled environment of different water treatments. We make use of a multispectral camera mounted on an Unmanned Aerial Vehicle for collecting the data from the tillering stage to the heading stage of the crop. A pre-processing pipeline, followed by the extraction of the Region of Interest from orthomosaic is explained. We propose a model based on a Convolution Neural Network, added with a deformable convolutional layer in order to learn and extract rich spatial and spectral features. These features are further fed to a weighted Attention-based Bi-Directional Long Short-Term Memory network to process the sequential dependency between temporal features. Finally, the water stress category is predicted using the aggregated Spatial-Spectral-Temporal Characteristics. The addition of multispectral, multi-temporal imagery significantly improved accuracy when compared with mono-temporal classification. By incorporating a deformable convolutional layer and Bi-Directional Long Short-Term Memory network with weighted attention, our proposed model achieved best accuracy of 91.30% with a precision of 0.8888 and a recall of 0.8857. The results indicate that multispectral, multi-temporal imagery is a valuable tool for extracting and aggregating discriminative spatial-spectral-temporal characteristics for water stress classification.

A novel algorithm for Bluetooth ECG.

In wireless transmission of ECG, data latency will be significant when battery power level and data transmission distance are not maintained. In applications like home monitoring or personalized care, to overcome the joint effect of previous issues of wireless transmission and other ECG measurement noises, a novel filtering strategy is required. Here, a novel algorithm, identified as peak rejection adaptive sampling modified moving average (PRASMMA) algorithm for wireless ECG is introduced. This algorithm first removes error in bit pattern of received data if occurred in wireless transmission and then removes baseline drift. Afterward, a modified moving average is implemented except in the region of each QRS complexes. The algorithm also sets its filtering parameters according to different sampling rate selected for acquisition of signals. To demonstrate the work, a prototyped Bluetooth-based ECG module is used to capture ECG with different sampling rate and in different position of patient. This module transmits ECG wirelessly to Bluetooth-enabled devices where the PRASMMA algorithm is applied on captured ECG. The performance of PRASMMA algorithm is compared with moving average and S-Golay algorithms visually as well as numerically. The results show that the PRASMMA algorithm can significantly improve the ECG reconstruction by efficiently removing the noise and its use can be extended to any parameters where peaks are importance for diagnostic purpose.

Early detection of breast cancer: synthesis and characterization of novel target specific NIR-fluorescent estrogen conjugate for molecular optical imaging.

Estrogen induced proliferation of existing mutant cells is widely understood to be the major risk determining factor in the development of breast cancer. Hence determination of the Estrogen Receptor[ER] status is of paramount importance. We have carried out the synthesis and characterization of a novel NIR fluorescent dye conjugate aimed at measuring ER+ve status in-vivo. The conjugate was synthesized by ester formation between 17-ß estradiol and a cyanine dye namely: bis-1, 1-(4-sulfobutyl) indotricarbocyanine-5-carboxylic acid, sodium salt. The replacement of the sodium ion in the ester by a larger glucosammonium ion was found to enhance the hydrophilicity and reduce the toxic effect on cell lines. The excitation and emission peaks for the dye were recorded as 750 and 788 nm respectively; ideal for non-invasive optical imaging owing to minimal tissue attenuation and auto-fluorescence at these wavelengths. The dye (NIRDC1) has a significant drop in plasma-protein binding therefore leading to marked improvement in pharmacokinetic profile such as dye evacuation in comparison to ICG. In addition the dye showed enhanced fluorescence quantum yield, molar extinction coefficient and linearity in fluorescence relative to ICG. This dye can be potentially used as a target specific exogenous contrast agent in molecular optical imaging for early detection of breast cancer.