kiran kumar chanumolu

Mr. kiran kumar chanumolu


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 December 2024 in Heliyon
Verified
Journal Article
Open access
MNMuni Nagamani
Muni Nagamani
KCkiran kumar chanumolu
kiran kumar chanumolu
Real-time monitoring and anomaly detection are essential in healthcare to ensure safe conditions for patients and maintain the integrity of medical data samples. The majority of existing systems, despite improvements in healthcare technologies, cannot capture the spatial and temporal patterns of multimodal data simultaneously, process high Volume data in real-time, and ensure the privacy of patients' identity effectively. In this work, we handle these limitations by proposing a complete approach that uses state-of-the-art deep learning and data processing architectures to realize resilient anomaly detection in healthcare systems. In this paper, we propose an advanced hybrid model for Convolutional and Long Short-Term Memory (CNN-LSTM), which exploits the main advantages of convoluted neural networks and LSTM networks. The proposed model will extract spatial features from medical images and temporal dependencies between them from patient vitals. These three components come together to give high accuracy in anomaly detection, where the accuracy rate was 94 % and a false positive rate as low as 2 % on a test dataset containing 10,000 patients. Furthermore, Apache Kafka will be integrated with TensorFlow Serving for real-time data processing to achieve scale. Conversely, Apache Kafka scales to high throughputs, providing low latencies in message processing, and TensorFlow Serving enables implementation of models into production at scale and serving predictions in real-time situations. This system processes over 100,000 messages per second with less than 50 ms of inference latency to enable the prompting of clinical responses. We use Federated Learning combined with Differential Privacy to solve our data security and privacy tasks. This will enable model training across multiple institutions without compromising the privacy of patient data samples. In the different use-case scenarios, local models update gradients to a central server while achieving model accuracy at 90 % with a privacy loss, epsilon, below 1 %. This will be used to implement a transformer-based multimodal integration model that will integrally fuse these diversified data types, including EHR text data, medical images, and time-series sensor data samples. This model improves the anomaly detection to an F1-score of 0.92, which means a performance increase of 15 % over all unimodal approaches. The proposed methods offer multifaceted solutions that provide advanced machine learning techniques, second-by-second real-time processing, and strict privacy measures all at once. This work significantly increases the reliability and security of healthcare systems, ensuring better patient outcomes while safeguarding sensitive medical data samples.

1 February 2025 in AIP Advances
Verified
Journal Article
Open access
KDK. Devipriya
K. Devipriya
MTMehdi Tlija
Mehdi Tlija
KCkiran kumar chanumolu
kiran kumar chanumolu
VKV. Chandra Kumar
V. Chandra Kumar
SJSubrata Jana
Subrata Jana
CJChiranjibe Jana
Chiranjibe Jana
Micro-plastics, which are plastic particles less than 5 mm in size, pose significant environmental threats due to their persistence and potential toxicity to marine life and humans. This work compares the effectiveness of several convolutional neural network (CNN) designs, including MobileNetV3Large, ResNet50V2, ResNet101V2, and EfficientNetB7, in identifying and categorizing microplastics into three groups: beads, fibers, and fragments. We evaluate the models using precision, recall, and F1-score criteria. The outcomes indicate that while all models perform well in identifying fiber microplastics, achieving high recall and moderate precision, they struggle significantly with bead and fragment categories. EfficientNetB7 and MobileNetV3Large exhibited the highest performance for fiber detection but failed to detect bead and fragment microplastics. The findings highlight the need for further research to enhance the classification accuracy for bead and fragment micro-plastics, suggesting that future work should focus on addressing class imbalance, utilizing advanced techniques such as transfer learning, and incorporating domain-specific knowledge to improve feature discrimination. This work provides the path for more efficient environmental monitoring systems by offering insightful information about the advantages and disadvantages of the CNN architectures now in use for the detection of microplastics.

28 February 2022 in Ingénierie des Systèmes d'information
Verified
Journal Article
Closed access
KCkiran kumar chanumolu
kiran kumar chanumolu
NRNandhakumar Ramachandran
Nandhakumar Ramachandran
With technological breakthroughs and increasing network technology reliability in daily life, it is vital to provide robust network-based system operations. According to the Comprehensive Error Rate Test Report, as network attacks become more common, the number of attacks nearly doubles or triples every year. The fifth-generation mobile communication is going to provide a connecting world with almost zero Latency. Network coding has emerged as a viable answer to the mostly efficiency, secure network needs for coming-generation networking technology. A small cell environment with network coding allows an effective system to communicate with the high rate of data. A significant countermeasure for several forms of network attacks is the Intrusion Detection Systems (IDS). Unique IDS solutions which are lightweight but also provide a high level of security are therefore demanded. The main security concerns in collaborative networks are Distributed -Denial- of- Service (DDOS) attacks, pollution attacks etc. While the network efficiency of non-DDOS attacks is also impaired, the impact of DDOS attacks is severe. In DDOS attacks, the specific node as a victim floods and mass traffic jams occurs, affecting the entire network performance. However, small, NC-enabled mobile cells, due to the essential NC vulnerabilities, are susceptible to pollution attacks. This article presents a brief survey on intrusion detection and mitigation schemes for Network Coding environment Mobile Small Cells.

31 October 2024 in IEEE Access
Verified
Journal Article
Open access
SMSpandana Mande
Spandana Mande
KCkiran kumar chanumolu
kiran kumar chanumolu
SBShaik Salma Asiya Begum
Shaik Salma Asiya Begum
Fernando MoreiraFernando Moreira
Fernando Moreira
The vehicular ad hoc networks VANET is an essential part of intelligent transportation systems (ITSs) since it may offer various multimedia services and safety services to pedestrians, passengers, and even drivers. A wireless communication protocol called dedicated short-range communication (DSRC) was created for toll collection systems. Nevertheless, DSRC standards are extremely constrained, necessitating the development of next-generation communication protocols appropriate for VANET. Here, intended to develop an Optimized Reinforcement Learning (ORL) for obtaining resource allocation in VANET. This proposed methodology is developed for achieving resource allocation with efficient data transmission. This proposed approach is utilized to adjust the control channel interval (CCI) and service channel interval (SCI) to empower network performance. Additionally, it is utilized to reduce data collisions and optimize the network’s backoff distribution. The proposed method is a combination of reinforcement learning (RL) and adaptive coati optimization (ACO). The RL is utilized to obtain an efficient channel access algorithm. In the RL, the Q value is optimally selected by using ACO. Based on this algorithm, the proposed method is utilized to enhance the performance of VANET data transmission by achieving optimal resource allocation. The proposed method is implemented in MATLAB, and performances are evaluated using performance measures. Additionally, to validate the performance of the proposed methodology, it is compared with conventional techniques.

1 December 2025 in Alexandria Engineering Journal
Verified
Journal Article
Open access
SKSahebgoud Karaddi
Sahebgoud Karaddi
VYVipin Prakash Yadav
Vipin Prakash Yadav
Pradeep Reddy GogulamudiPradeep Reddy Gogulamudi
Pradeep Reddy Gogulamudi
KCkiran kumar chanumolu
kiran kumar chanumolu
MSManumula Srinubabu
Manumula Srinubabu
AJAbhilasha Joshi
Abhilasha Joshi

8 December 2025 in Engineering, Technology & Applied Science Research
Verified
Journal Article
Open access
SMSreerama Murty Maturi
Sreerama Murty Maturi
CDC. Dastagiraiah
C. Dastagiraiah
PSPonnuru Sowjanya
Ponnuru Sowjanya
KCkiran kumar chanumolu
kiran kumar chanumolu
JSJ. S. V. R. S. Sastry
J. S. V. R. S. Sastry
Breast cancer survival rate can be dramatically affected if the disease is diagnosed early and the stage is classified accurately. Model-driven or statistical-driven Machine Learning (ML) techniques are powerful and reliable methods of diagnosing breast cancer, with the ability to utilize, sort, and detect patterns in data, and to handle immense databases. This research work considers the Breast Cancer Wisconsin (BCW) dataset to train several ML models (Random Forest, Support Vector Machines, Gradient Boosting, Logistic Regression, k-Nearest Neighbors) to classify images as either benign or malignant and compares the results. The followed methodology involves data preprocessing and feature extraction. All tested algorithms performed satisfactorily with Random Forest surpassing the others.