Taha Alhersh

Driving Digitalization and AI Innovations
BASF Digital Solutions

Medical Informatics

I have worked as Medical Computer Scientist on HiGHmed project to find novel and interoperable solutions in medical informatics.

Activity Recognition using Multi-modality Data

PhD research Data and Web Science Group focused on using machine learning and machine vision for activity recognition.

Metrics Performance Analysis of Optical Flow

In this research new optical flow performance metrics have been proposed and evaluated alongside with current metrics.

Unsupervised Fine-tuning of Optical Flow

An unsupervised loss function based on classical variational optical flow for deep learning based optical flow methods i.e. Flownet2-SD.

Action Recognition using Optical Flow and IMU

Using optical flow and IMU data as complementary to each other for better human bevaviour understanding.

During and after my Masters I was working on Machine Learning and Pattern Recognition fields for Medical Imaging and Biological Data.

Species Identification Using Machine Learning

We have developed an algorithm for species identification using part of DNA sequences. Classification approaches have been applied on different parts of the DNA sequences. Statistical features have been extracted and tested on various parts of the DNA sequences.

Classify Lung Cancer on Chest Radiographs

Introduced a CAD system that is comprised of feature extraction, feature selection and classification using a combination of wavelet and curvelet for feature extraction.

Source Code Classification

Using Latent Semantic Indexing classifier that integrates information on structural and frequency of terms in its weighting scheme for source code classification. The content terms are identified by extracting words in the source code program.

Research Interest: Past and Future

Multidisciplinary

Artificial Intelligence (AI) has been always of my interest. During my Masters I have been in touch with many AI subjects like: Data mining, Intelligent Databases, Intelligent Agent Development, Expert Systems, PROLOG, Artificial Neural Networks and other interesting AI subjects.

My interest in Bioinformatics started while doing my Masters thesis, in which I used statistical machine learning algorithms on Biological Data (DNA Sequences).

Three things have sparked my interest in Neuroscience and motivated me to be curious to think deeply in Human Brain: My eager to understand human behaviour and explor the human brain, brain inspired computations and especially Artificial Neural Network (ANN), and an inspiring video titled “Advice from a Nobel Laureate” which Prof. Susumu Tonegawa encouraged young people to direct their research towards Neuroscience which contains many mysteries need to be revealed.

Publications

  • Alhersh, Taha. (2021). From Motion to Human Activity Recognition. Dissertation, Mannheim.
  • Alhersh, T.; Belhaouari, S. and Stuckenschmidt, H. (2020). Metrics Performance Analysis of Optical Flow.In Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, ISBN 978-989-758-402-2, ISSN 2184-4321, pages 749-758.
  • Taha Alhersh, Samir Brahim Belhaouari, Heiner Stuckenschmidt(2019). Action Recognition Using Local Visual Descriptors and Inertial Data. In: Chatzigiannakis I., De Ruyter B., Mavrommati I. (eds) Ambient Intelligence. AmI 2019. Lecture Notes in Computer Science, vol 11912. Springer, Cham
  • Taha Alhersh and Heiner Stuckenschmidt. "On the Combination of IMU and Optical Flow for Action Recognition". In 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019 : Kyoto, Japan.
  • Taha Alhersh and Heiner Stuckenschmidt. "Unsupervised Fine-tuning of Optical Flow for Better Motion Boundary Estimation" In International Conference on Computer Vision Theory and Applications (VISAPP) 2019. Prague, Czech Republic.
  • Alhersh, Taha, Brahim Belhaouari Samir, Hamada Al-Absi, Abdullah Alorainy, and Belloui Bouzid. "Species identification using part of DNA sequence: Evidence from machine learning algorithms." In Proceedings of the 9th EAI International Conference on Bio-inspired Information and Communications Technologies (formerly BIONETICS), pp. 490-494. ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering), 2016.
  • Al-Absi, Hamada RH, Brahim Belhaouari Samir, Taha Alhersh, and Suziah Sulaiman. "On the combination of wavelet and curvelet for feature extraction to classify lung cancer on chest radiographs." In 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 3674-3677. IEEE, 2013.
  • Yusof, Yuhanis, Taha Alhersh, Massudi Mahmuddin, and Aniza Mohamed Din. "Source Code Classification using Latent Semantic Indexing with Structural and Frequency Term Weighting." Research Journal of Applied Sciences 7, no. 5 (2012): 266-271.

    • QOUTES

    So have they not traveled through the earth and have hearts by which to reason and ears by which to hear? For indeed, it is not eyes that are blinded, but blinded are the hearts which are within the breasts. [Quran 22, 6]

    Two things are infinite: the universe and human stupidity; and I'm not sure about the universe. - Albert Einstein

    And said, 'Ask forgiveness of your Lord. Indeed, He is ever a Perpetual Forgiver. (11) He will send [rain from] the sky upon you in [continuing] showers (12) And give you increase in wealth and children and provide for you gardens and provide for you rivers. [Quran, 71, 11-12]

    It is strange that only extraordinary men make the discoveries, which later appear so easy and simple. - Georg C. Lichtenberg

    Shall I refuse my dinner because I do not fully understand the process of digestion? - Oliver Heaviside

    Template from http://www.Styleshout.com