Mahbub Ul Alam

PhD Student Mahbub Ul Alam Received the Best Student Paper Award at the IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS 2022)

August 1, 2022 Mahbub Ul Alam Award, Research Paper, SAS

Hello Everyone,

Greetings. I hope you are well. I would like to share some very good news with you.

I recently published a paper with two other co-authors at the IEEE 35th International Symposium on Computer Based Medical Systems (CBMS 2022, July 21-23, 2022, Shenzhen, China, Online Event). CBMS is the premier conference for computer-based medical systems, and one of the main conferences within the fields of medical informatics and biomedical informatics.

The title of my paper is “Exploring LRP and Grad-CAM visualization to interpret multi-label-multi-class pathology prediction using chest radiography“, (Mahbub Ul Alam, Jón Rúnar Baldvinsson and Yuxia Wang)”. In this paper, we tried to explain the decision process of deep neural networks to predict pathology (abnormality) in chest-X ray data using two popular interpretable methods. We investigated whether this explanation matches the clinical diagnosis or not. Interpretability is very crucial and it is emphasized in the recent European Union Artifical Intelligence Act. We hope that this paper will create a positive impact in this aspect.

The paper was received well during the CBMS 2022 symposium presentation time. I am delighted to inform you that the paper received the ‘best student paper award’. The award was provided by the IEEE Technical Committee on Computational Life Science (TCCLS).

I am very honoured and would like to thank DSV for providing me with this opportunity. I am fortunate to be working here to get a second award for my research work. Previously I won the ‘best paper award’ at BIOSTEC HEALTHINF 2020 (you can read more about it here).

Want to know more about the paper? Please check out the following presentation video I made!

Abstract:

The area of interpretable deep neural networks has received increased attention in recent years due to the need for transparency in various fields, including medicine, healthcare, stock market analysis, compliance with legislation, and law. Layer-wise Relevance Propagation (LRP) and Gradient-weighted Class Activation Mapping (Grad-CAM) are two widely used algorithms to interpret deep neural networks. In this work, we investigated the applicability of these two algorithms in the sensitive application area of interpreting chest radiography images. In order to get a more nuanced and balanced outcome, we use a multi-label classification-based dataset and analyze the model prediction by visualizing the outcome of LRP and Grad-CAM on the chest radiography images. The results show that LRP provides more granular heatmaps than Grad-CAM when applied to the CheXpert dataset classification model. We posit that this is due to the inherent construction difference of these algorithms (LRP is layer-wise accumulation, whereas Grad-CAM focuses primarily on the final sections in the model’s architecture). Both can be useful for understanding the classification from a micro or macro level to get a superior and interpretable clinical decision support system.

PhD Student Mahbub Ul Alam Received the Best Paper Award at BIOSTEC HEALTHINF 2020

March 1, 2020 Mahbub Ul Alam Award, Research Paper

Hello Everyone,

Greetings. I hope you are well. I would like to share some very good news with you.

I presented a research paper entitled ‘Deep Learning from Heterogeneous Sequences of Sparse Medical Data for Early Prediction of Sepsis‘ at the 13th International Conference on Health Informatics (HEALTHINF), Malta, Valletta, 24-26 February 2020. HEALTHINF is part of BIOSTEC, the 13th International Joint Conference on Biomedical Engineering Systems and Technologies. It was organized by the Institute for Systems and Technologies of Information, Control and Communication (INSTICC).

The paper won the HEALTHINF 2020 Best Paper Award. The decision criterion considered both the paper quality (assessed by the Program Committee), and the oral presentation quality (feedback given by session chairs at the conference venue). Overall it was an enjoyable experience, and I learned a lot from all.

Want to know more about the paper or sepsis in general? Please check out the following presentation video I made!

Abstract:

Sepsis is a life-threatening complication to infections, and early treatment is key for survival. Symptoms of sepsis are difficult to recognize, but prediction models using data from electronic health records (EHRs) can facilitate early detection and intervention. Recently, deep learning architectures have been proposed for the early prediction of sepsis. However, most efforts rely on high-resolution data from intensive care units (ICUs). Prediction of sepsis in the non-ICU setting, where hospitalization periods vary greatly in length and data is more sparse, is not as well studied. It is also not clear how to learn effectively from longitudinal EHR data, which can be represented as a sequence of time windows. In this article, we evaluate the use of an LSTM network for early prediction of sepsis according to Sepsis-3 criteria in a general hospital population. An empirical investigation using six different time window sizes is conducted. The best model uses a two-hour window and assumes data is missing not at random, clearly outperforming scoring systems commonly used in healthcare today. It is concluded that the size of the time window has a considerable impact on predictive performance when learning from heterogeneous sequences of sparse medical data for early prediction of sepsis.

Best Paper Award Certificate

Receiving Award from the Program Chair Dr. Fedrico Cabitza

2018 Conference on Empirical Methods in Natural Language Processing (EMNLP 2018)

November 13, 2018 Mahbub Ul Alam Health Informatics

Hello everyone. I hope you are all fine and enjoying this beautiful winter. 🙂

Recently I have attended the EMNLP 2018 conference. It was held in Brussels, Belgium at the city center from 31 October to 04 November.

Just a few statistics about the conference,

EMNLP 2018 had 14 workshops, six tutorials, three invited speakers, 351 long paper presentations, 198 short paper presentations, 10 TACL paper presentations, and 29 demos. It received 2,231 valid submissions, a 48% increase over EMNLP 2017.

I attended the ”LOUHI 2018”, the Ninth International Workshop on Health Text Mining and Information Analysis and the BlackboxNLP: Analyzing and interpreting neural networks for NLP workshop. The ”LOUHI 2018”, the health text mining and information analysis workshop provided an excellent overview of the current trend in this area. The total number of attendee was close to a hundred. I have observed that most of the works there are based on deep neural networks, and they have particular importance on interpretability. The BlackboxNLP workshop was focused on this interpretability issue, and I have learned some recent trends there. I have written a short report about some papers of the the conference that can be sent on request.

Brussels is a magnificent city. During the lunchtime, I tried to visit as much as possible the city center. I tried the Belgian chocolate, french fries, and the famous waffle. They were mouth watering delicious.

Overall it was an excellent experience for me. I met a lot of new people who are working in the same direction as mine. I learned some new concepts. Most importantly, it gave me immense strength and assurance that, I am not alone in this journey. There are a lot of researchers in this area. I can always learn from them.

Mahbub Ul Alam, A new employee

September 7, 2018 Mahbub Ul Alam Employee, Health Informatics, Immersive Participation, SAS, Ubiquitous Computing

Hello everyone! I hope you all are having a lovely time. 🙂

I am Mahbub Ul Alam. I’ve recently started as a PhD student here at DSV.

My research interest is in ‘Cognitive internet of things (IoT) based smart health-care systems using deep learning and advanced machine learning’.

Potential research problem areas include:
1. Machine learning framework for IoT.
2. Intelligent data, i.e., machine-understandable, resource-recognition, knowledge representation, big data, deep learning, and advanced machine learning.
3. Clinical decision support systems in health-IoT.
4. Patient-centric personalized health-care systems and applications.
5. Distributed intelligent data processing in IoT.

I have a master’s degree at the Institute of Natural Language Processing (IMS), University of Stuttgart, Germany. In my master’s thesis, I worked to understand the hidden layer mechanisms of deep neural networks in natural language processing (automatic speech recognition) domain.

I have a bit of work experience in software engineering at Samsung Research and Development Institute Bangladesh.

I believe in diversity, and love to explore new and fresh technological innovations. I like to think that every aspect of my previous experiences is helping me to move forward for my future career. I always keep in mind that, the central principle of my life is ‘Let all of us prosper together.’

The people are super nice and friendly here, and I am looking forward to working together with all of you at DSV.

Let all of us prosper together. 🙂

Mahbub

Mahbub Ul Alam