BISPL-Biomedical Instrumentation and Signal Processing Lab

Alhamdulillah! Two conference papers have been accepted at the 2025 5th International Conference on Robotics, Automation, and Artificial Intelligence (RAAI 2025) to be held on 18-20 Dec, 2025 in Singapore. Congratulations to all the coauthors involved.

Three members of BISPL-Biomedical Instrumentation and Signal Processing Lab, Tasfia Hasan Faiza (ID 2010245), Muhammad Sajid Hossain (ID 2030647), and Samara Islam (ID 2120309), received the 𝗜𝗻𝘀𝗽𝗶𝗿𝗶𝗻𝗴 𝗔𝘄𝗮𝗿𝗱 for the Outstanding Research Achievement in 2025 from the Department of EEE, Independent University, Bangladesh, during Fresher’s Fiesta, Autumn 2025.

Their research contributions are as follows (outcome of FYDP EEE400, supervised by Lab Head Dr. Md. Kafiul Islam):

1) Journal Publication: Tasfia Hasan Faiza, Muhammad Sajid Hossain, Samara Islam, Nazmus Sakib, Tasnuva Faruk, Md. Kafiul Islam, “Raw fetal PCG dataset contaminated with Mother’s PCG”- Data in Brief, Volume 63, 2025, 112107, ISSN 2352-3409, https://doi.org/10.1016/j.dib.2025.112107. (Q3 ranked Journal, WoS IF =1.2).

2) Conference Publication: Tasfia Hasan Faiza, Muhammad Sajid Hossain, Samara Islam, Nazmus Sakib, Tasnuva Faruk, and Md Kafiul Islam, “Fetal Heart Rate Extraction from a Noisy Maternal-Fetal Heart Sound Using EMD and Wavelet Transform” - accepted and presented at the 1st 2025 IEEE International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN 2025), BAUST, Saidpur. (indexed in IEEE Xplore and Scopus).

3) Dataset Publication: Hasan Faiza, Tasfia; Hossain, Muhammad Sajid ; Islam , Samara ; Sakib , Nazmus ; Faruk , Tasnuva ; Islam, Md Kafiul (2025), “Raw Fetal PCG Contaminated With Mother’s PCG”, Mendeley Data, V1, doi: 10.17632/k5z7hf6vbb.1

Lab Co-Head Dr. Tasnuva Faruk and BISPL Senior Member, Mr. Nazmus Sakib, were also involved in this research as mentors.

Their dedication and contributions continue to inspire the next generation of engineers. Wishing them more success ahead.

Congratulations, Tasfia Hasan Faiza, Muhammad Sajid Hossain, and Samara Islam!

Alhamdulillah. Another EEE400 FYDP outcome has been published in IEEE QPAIN 2025 Conference and is now available online in the IEEE Xplore Digital Library.

Paper Title: Fetal Heart Rate Extraction from a Noisy Maternal-Fetal Heart Sound Using EMD and Wavelet Transform

Authored by: Tasfia Hasan Faiza; Muhammad Sajid Hossain; Samara Islam; Nazmus Sakib; Tasnuva Faruk; Md. Kafiul Islam

Conference: 2025 International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN)

Conference Location: Rangpur, Bangladesh

DOI: 10.1109/QPAIN66474.2025.11172077

https://ieeexplore.ieee.org/document/11172077

Alhamdulillah. We have a new article titled "Raw Fetal PCG Dataset Contaminated with Mother’s PCG" published in Data in Brief of Elsevier. If you are interested, you can download the open-access article from the following link.
https://www.sciencedirect.com/science/article/pii/S2352340925008285
The PCG dataset can also be openly accessed through https://data.mendeley.com/datasets/k5z7hf6vbb/1

This is an outcome of EEE400: Final Year Design Project by Tasfia Hasan Faiza, Muhammad Sajid Hossain, Samara Islam, who are co-mentored by BISPL-Biomedical Instrumentation and Signal Processing Lab Senior Member Mr. Nazmus Sakib and Lab Co-head Dr. Tasnuva Faruk.

𝗡𝗲𝘂𝗿𝗮𝗹𝗶𝗻𝗸 𝗨𝗽𝗱𝗮𝘁𝗲, 𝗦𝘂𝗺𝗺𝗲𝗿 𝟮𝟬𝟮𝟱:

https://youtu.be/FASMejN_5gs?si=PkPTW2HXnAnKJ2xs

00:00 Countdown
00:30 Intro & Overview (Elon)
12:39 Progress & Roadmap (DJ)
24:19 Our Participants (Sehej)
32:41 Robot Control & Body Reanimation (Elon)
35:07 Brain Computer Interfaces (Nir)
37:43 User Experience (Rooz)
40:32 Neural Decoding (Harrison)
45:02 Blindsight (Joey)
49:29 Robot (John)
53:14 Implant (Julian)
58:21 Closing (DJ & Elon)

Neuralink Update, Summer 2025 We’re translating neural signals into life-changing impact.Join us to build the machine that builds the machine to merge with machines.neuralink.com/careersL...

The IEEE Computer Society Bangladesh Chapter Summer Symposium was held on July 18–19, 2025, at Hajee Mohammad Danesh Science & Technology University, Dinajpur.

We are pleased to share that Partho Prosad, Research Assistant at BISPL-Biomedical Instrumentation and Signal Processing Lab, presented his research Extended Abstract as the first author at the event.

Research Title: “𝙏𝙝𝙚 𝙀𝙛𝙛𝙚𝙘𝙩 𝙤𝙛 𝘼𝙧𝙩𝙞𝙛𝙖𝙘𝙩 𝙍𝙚𝙢𝙤𝙫𝙖𝙡 𝙛𝙧𝙤𝙢 𝙀𝙀𝙂 𝙎𝙞𝙜𝙣𝙖𝙡𝙨 𝙤𝙣 𝙩𝙝𝙚 𝘿𝙚𝙩𝙚𝙘𝙩𝙞𝙤𝙣 𝙤𝙛 𝙀𝙥𝙞𝙡𝙚𝙥𝙩𝙞𝙘 𝙎𝙚𝙞𝙯𝙪𝙧𝙚𝙨.”

This work was part of his EEE400 Final Year Design Project, supervised by Dr. Md. Kafiul Islam, Associate Professor at the Department of EEE, Independent University, Bangladesh, and Head of the BISPL-Biomedical Instrumentation and Signal Processing Lab.

We are proud of this achievement and look forward to more impactful work from the BISPL team!

Symposium: ss25.ieeecsbdc.org
Abstract: dx.doi.org/10.13140/RG.2.2.24715.20005
Presentation: dx.doi.org/10.13140/RG.2.2.18141.12008

The Department of Computer Science and Engineering at Independent University, Bangladesh (IUB), organized the 2nd CSE Annual Research Day (CARD) 2025. This event celebrates research, innovation, and student success.

We are happy to share that Mobarak Hossen Mollah Emon, a Research Assistant at BISPL-Biomedical Instrumentation and Signal Processing Lab, won the 𝐁𝐞𝐬𝐭 𝐒𝐭𝐮𝐝𝐞𝐧𝐭 𝐑𝐞𝐬𝐞𝐚𝐫𝐜𝐡𝐞𝐫 𝐀𝐰𝐚𝐫𝐝 in the Engineering & Physical Sciences category.

Research Title: “𝙈𝙖𝙘𝙝𝙞𝙣𝙚 𝙇𝙚𝙖𝙧𝙣𝙞𝙣𝙜 𝘽𝙖𝙨𝙚𝙙 𝙉𝙚𝙪𝙧𝙤𝙢𝙪𝙨𝙘𝙪𝙡𝙖𝙧 𝘿𝙞𝙨𝙚𝙖𝙨𝙚 𝘿𝙚𝙩𝙚𝙘𝙩𝙞𝙤𝙣 𝙖𝙣𝙙 𝘾𝙡𝙖𝙨𝙨𝙞𝙛𝙞𝙘𝙖𝙩𝙞𝙤𝙣 𝙐𝙨𝙞𝙣𝙜 𝙀𝙈𝙂 𝙎𝙞𝙜𝙣𝙖𝙡.”

This research was part of his EEE400 Final Year Design Project, supervised by Dr. Md. Kafiul Islam, Associate Professor and Head of BISPL.

Co-authors: Ariful Hasan Akash, Tahmid Ahmed Bhuiyan, Tazin Sharin, Nazmus Sakib, and Tasnuva Faruk.

Read the paper: https://ieeexplore.ieee.org/document/10914752

We are proud of this achievement and look forward to more great work from the BISPL team!

We have uploaded two sets of open-source EEG datasets for Depression and Anxiety Screening, respectively, which were recorded by our volunteers at the BISPL-Biomedical Instrumentation and Signal Processing Lab. Interested researchers are invited to download and test their signal processing and ML/DL algorithms on these datasets, sharing their feedback with us.
-------------------------------------------------------------
Dataset-1:
Multi-channel Wireless EEG Recordings of Young Adults for Depression Screening based on PHQ-9 https://data.mendeley.com/datasets/ct5vrc4k2t/1

The EEG was recorded using a wireless EMOTIV EPOC+ headset (with 14 channels and a sampling rate of 128 Hz) from 31 young adults (aged between 18 and 25 years, 15 male and 16 female). Before EEG acquisition, after taking participant consent, a self-reported survey following the PHQ-9 questionnaire was filled out by the participants to find the ground truth for screening depression. There were 18 participants found to have PHQ-9 scores more than or equal to 20 classified as Depressed subjects (labelled as DSub1-DSub18), while 13 participants had PHQ-9 scores less than or equal to 4 classified as Depression Control subjects (labelled as DCSub1-DSub13). Each recording was 5 minutes long for each participant. The 14 EEG channels are placed according to the International 10-20 electrode montage system: eight frontal electrodes (AF3, F3, F7, FC5, AF4, F4, F8 and FC6), two temporal electrodes (T7 and T8), two parietal electrodes (P7 and P8), two occipital electrodes (O1 and O2), and two reference channels (P3 and P4). The dataset has .mat file extension (can be opened by MATLAB software). Each file has a data size of 38,400 x 14, where each column denotes channel number and each row denotes sample number. Since each recording is 5 minutes long (300 seconds), each channel has 38,400 samples, which is equivalent to 300 seconds (sampling rate of 128 Hz).

Please cite the following articles if you use this dataset:

1. Sakib, Nazmus, Md Kafiul Islam, and Tasnuva Faruk. "Machine learning model for computer‐aided depression screening among young adults using wireless EEG headset." Computational Intelligence and Neuroscience 2023, no. 1 (2023): 1701429.

2. N. Sakib, M. K. Islam and T. Faruk, "Machine Learning Based Depression Screening Among Young Adults Using Wireless EEG," 2023 International Conference on Artificial Intelligence Innovation (ICAII), Wuhan, China, 2023, pp. 110-115, doi: 10.1109/ICAII59460.2023.10497265.

3. Sakib, Nazmus, Md Kafiul Islam, and Tasnuva Faruk. "Effect of Artifact Removal in Machine Learning Based Depression Screening using EEG." In Proceedings of the 2023 8th International Conference on Biomedical Imaging, Signal Processing, pp. 115-120. 2023.

---------------------------------------------------------------------

Dataset-2:
Multi-channel Wireless EEG Recordings of Young Adults for Anxiety Screening based on GAD-7

https://data.mendeley.com/datasets/rh3fy75zdv/1

The EEG was recorded using a wireless EMOTIV EPOC+ headset (with 14 channels and a sampling rate of 128 Hz) from 38 young adults (aged between 18 and 25 years). Before EEG acquisition, after taking participant consent, a self-reported survey following the GAD-7 questionnaire was filled out by the participants to find the ground truth for screening anxiety. There were 23 participants found to have GAD-7 scores more than or equal to 15 classified as Anxiety (labelled as ASub1-ASub23), while 15 participants had GAD-7 scores less than or equal to 4 classified as Anxiety Control subjects (labelled as ACSub1-ACSub15). Each recording was 5 minutes long for each participant. The 14 EEG channels are placed according to the International 10-20 electrode montage system: eight frontal electrodes (AF3, F3, F7, FC5, AF4, F4, F8 and FC6), two temporal electrodes (T7 and T8), two parietal electrodes (P7 and P8), two occipital electrodes (O1 and O2), and two reference channels (P3 and P4). The dataset has .mat file extension (can be opened by MATLAB software). Each file has a data size of 38,400 x 14, where each column denotes channel number and each row denotes sample number. Since each recording is 5 minutes long (300 seconds), each channel has 38,400 samples, which is equivalent to 300 seconds (sampling rate of 128 Hz).

Please cite the following articles if you use this dataset:

1) Sakib, Nazmus, Tasnuva Faruk, and Md Kafiul Islam. "Wireless EEG based anxiety screening among young adults using machine learning model." In Proceedings of the 2023 8th International Conference on Biomedical Imaging, Signal Processing, pp. 97-103. 2023.

2) N. Sakib, K. Islam and T. Faruk, "Effect of Artifact Removal in Machine Learning Based Anxiety Screening Using EEG Signal," 2025 4th International Conference on Robotics, Electrical and Signal Processing Techniques (ICREST), Dhaka, Bangladesh, 2025, pp. 498-503, doi: 10.1109/ICREST63960.2025.10914438.
-------------------------------------------------------------------
In both cases, the ethical approval through the Institutional Review Board (IRB) of the Independent University, Bangladesh (IUB) was taken prior to the experiments. All the participants were students of IUB, whose EEG recordings were conducted at the Biomedical Instrumentation and Signal Processing Lab (BISPL) of the Department of Electrical and Electronic Engineering (EEE), IUB.

We have uploaded an Open-source Dataset on Mendeley Data website titled "Raw Fetal PCG Contaminated With Mother’s PCG", which was recorded by FYDP students during their project work with the BISPL-Biomedical Instrumentation and Signal Processing Lab. Interested researchers are invited to download and test their signal processing and ML/DL algorithms on the dataset, sharing their feedback with us.

Download link: https://data.mendeley.com/datasets/k5z7hf6vbb/1

Citation: Hasan Faiza, Tasfia; Hossain, Muhammad Sajid ; Islam , Samara ; Sakib , Nazmus ; Faruk , Tasnuva ; Islam , Md Kafiul (2025), “Raw Fetal PCG Contaminated With Mother’s PCG”, Mendeley Data, V1, doi: 10.17632/k5z7hf6vbb.1

Data Description: Eight patients consented in the participation of the recording for the fetus PCG. Most of the patients were 36+ weeks pregnant except one patient with a pregnancy of 32+ weeks. The hardware used to record fetal PCG is non-invasive with the infrastructure of a stethoscope attached to the microphone. Doppler and pulse oximeter along with the hardware were used simultaneously to record fetal and mother PCG. The sampling rate of the PCG is 44100 Hz. The doppler and pulse oximeter are the standard device used to record the fetus heart rate and mother’s heart rate in beats per minute (bpm). In contrast, the PCG signals acquired acoustic fetal heart sound traces. Each patient was in laying position while the experiment was carried out. The recordings were taken in three and two sessions per patient in two batches. The first batch consists of 3 patients with 2 sessions per minute. And the second batch includes 5 patients with 3 sessions per minute. The naming of the files is done accordingly such that B1_P1_S1 represents Batch 1’s patient 1 during session 1. And B2_P2_S2 indicates Batch 2’s patient 2 during session 2. Each session lasted for 1 minute with a gap of 10 seconds per session for both batches. The hardware was assembled under the Biomedical Instrumentation and Signal Processing Laboratory (BISPL) at Independent University, Bangladesh. The fetal and maternal PCG data were recorded in Urban Maternity Center, Dhaka North City Corporation, Bangladesh.

Dr. Md Kafiul Islam, Associate Professor in the Department of EEE, IUB at Independent University, Bangladesh, and Head of the BISPL-Biomedical Instrumentation and Signal Processing Laboratory, will deliver an invited talk, titled "𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗠𝗼𝗱𝗲𝗹 𝗳𝗼𝗿 𝗔𝗻𝘅𝗶𝗲𝘁𝘆 𝗦𝗰𝗿𝗲𝗲𝗻𝗶𝗻𝗴 𝗮𝗺𝗼𝗻𝗴 𝗬𝗼𝘂𝗻𝗴 𝗔𝗱𝘂𝗹𝘁𝘀 𝗨𝘀𝗶𝗻𝗴 𝗪𝗶𝗿𝗲𝗹𝗲𝘀𝘀 𝗘𝗘𝗚 𝗛𝗲𝗮𝗱𝘀𝗲𝘁," on 25 May 2025, 12:00–12:40 PM (Dhaka time), ONLINE, at the 𝟮𝟬𝟮𝟱 𝗜𝗘𝗘𝗘 𝟭𝟰𝘁𝗵 𝗜𝗻𝘁𝗲𝗿𝗻𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗖𝗼𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲 𝗼𝗻 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀, 𝗖𝗶𝗿𝗰𝘂𝗶𝘁𝘀 𝗮𝗻𝗱 𝗦𝘆𝘀𝘁𝗲𝗺𝘀 (𝗜𝗖𝗖𝗖𝗔𝗦), to be held in Wuhan, China, from May 23–25, 2025.

Dr. Islam is also serving as a Special Session Editor and Chair for the session "𝗕𝗶𝗼𝗺𝗲𝗱𝗶𝗰𝗮𝗹 𝗜𝗻𝘀𝘁𝗿𝘂𝗺𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗦𝗶𝗴𝗻𝗮𝗹 𝗣𝗿𝗼𝗰𝗲𝘀𝘀𝗶𝗻𝗴 𝗳𝗼𝗿 𝗗𝗶𝘀𝗲𝗮𝘀𝗲 𝗗𝗶𝗮𝗴𝗻𝗼𝘀𝗶𝘀, 𝗣𝗿𝗼𝗴𝗻𝗼𝘀𝗶𝘀, 𝗮𝗻𝗱 𝗧𝗿𝗲𝗮𝘁𝗺𝗲𝗻𝘁."

Interested people are welcome to attend the Invited Talk Online: Meeting ID 849 7188 3091

Details are given in the links below:
ICCCAS 2025 Conference: https://www.icccas.org/
Special Session Details: https://www.icccas.org/SS7.html
Invited Speaker Details: https://www.icccas.org/SS7-IS-Md-Kafiul-Islam.html

We have uploaded two sets of open-source EEG datasets for MI-BCI applications, which were recorded by our volunteers at the BISPL-Biomedical Instrumentation and Signal Processing Lab. Interested researchers are invited to download and try & test their signal processing and ML/DL algorithms on these datasets and share their feedback with us.
-------------------------------------------------------------
EEG MI-BCI Mouse Cursor Dataset-1 (Left and Right Hand Movements) Binary Class: https://data.mendeley.com/datasets/zgtryg9wvg/2

This dataset contains EEG recordings collected for research on brain-computer interface (BCI) applications, specifically for controlling a mouse cursor using motor imagery (MI). The data was recorded using an Emotiv Epoc+ 14 channel headset. Volunteers performed 1) left-hand movement, 2) right-hand movement and 3) break tasks. The data were recorded in Biomedical Instrumentation and Signal Processing Laboratory (BISPL) at Independent University, Bangladesh.

Please cite the following article if you use this dataset:
S. T. Roja, S. B. Rafique, M. A. Rhaman, N. Sakib and M. K. Islam, "EEG-based Mouse Cursor Control using Motor Imagery Brain-Computer Interface," 2024 6th International Conference on Electrical Engineering and Information & Communication Technology (ICEEICT), Dhaka, Bangladesh, 2024, pp. 1042-1047, doi: 10.1109/ICEEICT62016.2024.10534379.

---------------------------------------------------------------------
EEG MI-BCI Multidirectional Mouse Cursor Dataset-2 (Left and Right Hand and Leg Movements) 4 Classes:
https://data.mendeley.com/datasets/c92c2n5t34/2

This dataset contains EEG recordings for a multidirectional motor imagery-based brain-computer interface (MI-BCI) designed for mouse cursor control. The data was collected using the EMOTIV EPOC+ 14-channel wireless EEG headset, capturing brain activity related to voluntary motor imagery tasks. Volunteers have performed trials for 1) right-hand movement, 2) right-leg movement, 3) left-hand movement, 4) left-leg movement, 5) eye blinks, and 6) resting state (break periods). The data were recorded in Biomedical Instrumentation and Signal Processing Laboratory (BISPL) at Independent University, Bangladesh.

What Happens in Your Brain When You Sleep?

Sleep isn’t just about resting—it’s a time when your brain works in amazing ways! Every night, your brain changes and creates dreams, which can affect how you think, feel, and live during the day. For biomedical engineers, this is a big deal because it helps us build better tools to study sleep and improve it.

Here’s how we can use this knowledge:
Better Brain Tests: Improve tools like EEGs to understand sleep and dreams.
Smart Sleep Gadgets: Create devices to help people sleep better.
Computer Models: Use tech to study how the brain works during dreams.

This video explains how sleep and dreams work and why they matter.
👉 Watch here: https://youtu.be/wysnmEyoyB0?si=m4LTypwUD1Lk07CD