Susha Cheriyedath

Feature Writer

B.Sc Chemistry, M.Sc. Biochemistry

Susha is a scientific communication professional holding a Master's degree in Biochemistry, with expertise in Microbiology, Physiology, Biotechnology, and Nutrition. After a two-year tenure as a lecturer from 2000 to 2002, where she mentored undergraduates studying Biochemistry, she transitioned into editorial roles within scientific publishing. She has accumulated nearly two decades of experience in medical communication, assuming diverse roles in research, writing, editing, and editorial management.

Passionate about health and nutrition, Susha is interested in various diet types, such as veganism, and their impact on the body and metabolism. She remains dedicated to staying current with the latest developments in the medical and health science fields. In her leisure time, she engages in gardening, cooking/baking, and sketching/painting. When not immersed in work, she can often be found tending to her garden (weather permitting) or engrossed in her sketching and painting projects.

See less

Articles from Susha

Human vs. Machine Facial Expression Recognition: A Study on Personality and Observational Behavior

This study investigated how humans recognize facial expressions using limited facial landmarks, similar to techniques in machine learning. The research revealed that personality traits significantly influence the accuracy of facial expression recognition, and restricting observational behaviors can impact the ability to recognize negative expressions. These findings shed light on the limitations of human recognition and offer insights for improving facial expression recognition applications in various fields.

6 Oct 2023

Make-An-Animation: Generating Human Motions from Textual Descriptions

Researchers introduce "Make-An-Animation," a model that generates lifelike human motions from textual descriptions. The model leverages large-scale image-text datasets containing 35 million pairs of textual descriptions and static poses to learn and generate static 3D body poses.

5 Oct 2023

PGPNet: Revolutionizing Pill Detection for Improved Medication Safety

Researchers introduce PGPNet, a groundbreaking multi-pill detection framework that addresses the issue of pill misuse by accurately identifying and localizing pills with visual similarities. This innovative approach utilizes a priori graphs and external knowledge to enhance detection precision, offering a promising solution to the problem of drug misuse and prescription errors.

2 Oct 2023

BlinkLinMulT: A Transformer-Based System for Efficient Eye Blink Detection

Researchers introduce BlinkLinMulT, a transformer-based system for detecting eye blinks effectively. This innovative approach combines diverse inputs, including RGB texture data, iris and eye landmarks features, and head pose angles, using a multimodal transformer with linear attention to achieve robust and accurate blink detection, even in challenging real-world scenarios.

28 Sep 2023

Advancing Urban Acoustic Management through Binaural Sensing and Cloud Processing

This article explores a recent research paper that introduces an innovative approach to urban noise monitoring by combining binaural sensing and cloud-based data processing. The proposed system utilizes a 3D-printed artificial head equipped with microphones to capture acoustic data, enabling more accurate and comprehensive noise analysis. The cloud-based architecture further processes the data, offering valuable spatial indicators for urban soundscape evaluations, thereby contributing to enhanced urban planning strategies and overall quality of life.

16 Aug 2023

Applications of AI in Autonomous Vehicles

AI plays a vital role in enabling autonomous vehicles (AVs) across various industries. This article explores the applications of AI in AVs, including perception, localization, and planning. It delves into the challenges faced, such as data requirements, sensor fusion, behavior modeling, and deep learning limitations, highlighting the need for further research and development to advance the deployment of AVs in real-world scenarios.

25 Jul 2023

Safeguarding IoT: Intelligent Anomaly Detection with Federated Learning and Machine Learning

The study proposes a smart system for monitoring and detecting anomalies in IoT devices by leveraging federated learning and machine learning techniques. The system analyzes system call traces to detect intrusions, achieving high accuracy in classifying benign and malicious samples while ensuring data privacy. Future research directions include incorporating deep learning techniques, implementing multi-class classification, and adapting the system to handle the scale and complexity of IoT deployments.

6 Jul 2023