Tabassum, S., Abedin, N., Rahman, M. M., Rahman, M. M., Ahmed, M. T., Islam, R., & Ahmed, A. (2022). An online cursive handwritten medical words recognition system for busy doctors in developing countries for ensuring efficient healthcare service delivery. Scientific reports, 12(1), 3601.
Tabassum, S., Ullah, S., Al-Nur, N. H., & Shatabda, S. (2020). Poribohon-BD: Bangladeshi local vehicle image dataset with annotation for classification. Data in brief, 33.
Tabassum, S., & Amirshahi, S. A. (2024, June). Quality of NeRF Changes with the Viewing Path an Observer Takes: A Subjective Quality Assessment of Real-time NeRF Model. In 2024 16th International Conference on Quality of Multimedia Experience (QoMEX) (pp. 88-91). IEEE.
Tabassum, S., Abedin, N., Maruf, R. I., Ahmed, M. T., & Ahmed, A. (2022, March). Improving health status prediction by applying appropriate missing value imputation technique. In 2022 IEEE 4th Global Conference on Life Sciences and Technologies (LifeTech) (pp. 345-348). IEEE.
Tabassum, S., Takahashi, R., Rahman, M. M., Imamura, Y., Sixian, L., Rahman, M. M., & Ahmed, A. (2021, May). Recognition of doctors’ cursive handwritten medical words by using bidirectional LSTM and SRP data augmentation. In 2021 IEEE Technology & Engineering Management Conference-Europe (TEMSCON-EUR) (pp. 1-6). IEEE.
Podder, K. K., Tabassum, S., Khan, L. E., Salam, K. M. A., Maruf, R. I., & Ahmed, A. (2021, May). Design of a sign language transformer to enable the participation of persons with disabilities in remote healthcare systems for ensuring universal healthcare coverage. In 2021 IEEE Technology & Engineering Management Conference-Europe (TEMSCON-EUR) (pp. 1-6). IEEE.
Tabassum, S., Sampa, M., Maruf, R., Yokota, F., Nakashima, N., & Ahmed, A. (2020). An analysis on remote healthcare data for future health risk prediction to reduce health management cost. In APAMI 2020c: 11th Biennial Conference of the Asia-Pacific Association for Medical Informatics (Vol. 115, p. 119).
Tabassum, S., Sampa, M. B., Islam, R., Yokota, F., Nakashima, N., & Ahmed, A. (2020, November). A data enhancement approach to improve machine learning performance for predicting health status using remote healthcare data. In 2020 2nd International Conference on Advanced Information and Communication Technology (ICAICT) (pp. 308-312). IEEE.
Tabassum, S., Ullah, M. S., Al-Nur, N. H., & Shatabda, S. (2020, June). Native vehicles classification on Bangladeshi roads using CNN with transfer learning. In 2020 IEEE Region 10 Symposium (TENSYMP) (pp. 40-43). IEEE.
Tabassum, S., Zaman, M. I. U., Ullah, M. S., Rahaman, A., Nahar, S., & Islam, A. M. (2019, December). The cardiac disease predictor: IoT and ML driven healthcare system. In 2019 4th International Conference on Electrical Information and Communication Technology (EICT) (pp. 1-6). IEEE.
Zaman, M. I. U., Tabassum, S., Ullah, M. S., Rahaman, A., Nahar, S., & Islam, A. M. (2019, May). Towards IoT and ML driven cardiac status prediction system. In 2019 1st International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT) (pp. 1-6). IEEE.
Predicting Cardiac Anomaly using IoT sensors and Machine Learning
Collect vital signs such as heart rate, electrocardiogram, blood pressure, blood cholesterol from the human body using sensor-based wearable technology.
Trained a neural network on Cleveland dataset from the UCI Machine Learning Repository to detect and notify cardiac abnormality of a patient. The model achieved 82% accuracy.
Developed a mobile app to connect with the sensors and notify the user in case any anomaly is found.
Bangladeshi Local Vehicles Detection and Classification Using CNN with Transfer Learning
Developed a perception system to detect and classify fifteen Bangladeshi native vehicles such as rickshaw, easybike, cng, van, boat, truck, tractor, horsecart.
Implemented Convolutional Neural Network (CNN) using TensorFlow.
Used CFG files and pre-trained weights of 'You Only Look Once (YOLO)' algorithm to build the CNN network.
Collected over 9000 images of fifteen vehicles and introduced 'Bangladeshi Vehicles Dataset'.
Wrote corresponding annotation files of the images in XML using LabelImg tool.
Applied Transfer Learning and trained the model for 100 epoches.
Achieved 73% IoU and 3.19 average loss.
Technology stack: YOLO algorithm, TensorFlow, DarkFlow, Transfer Learning.
Real-time Three-wheeler Local Vehicle ‘Rickshaw’ Detection Using CNN Algorithm
Developed a real-time object detection system to recognize a local three-wheeler vehicle 'Rickshaw' using the YOLO algorithm.
Implemented Convolutional Neural Network using TensorFlow.
Used CFG files and pre-trained weights of YOLO to build TensorFlow network.
Collected 1000 images of the vehicle from Bangladeshi roads.
Wrote corresponding annotation files of the images in XML using LabelImg tool.
Trained the model for 50 epochs and achieved 86% accuracy.
Technology stack: YOLO algorithm, TensorFlow, DarkFlow.