NAAMII Call for Internships & Training 2024
Are you interested in gaining practical experience in cutting-edge research and technology? Do you have a passion for solving complex problems in deep learning and medical imaging? NAAMII is pleased to announce its Internship & Training Program for the year 2024. This opportunity offers an exciting chance for talented individuals to work on impactful projects under the guidance of experienced supervisors and researchers. Position 1: Operator-agnostic Abdominal Circumference Ultrasound Measurement Supervisor: Bishesh Khanal, Safal Thapaliya Summary Ultrasound is a low-cost medical imaging modality providing excellent visualization of internal body structures without the adverse effects of radiation. Hence, it is a preferred modality for diagnosing pregnancy-related abnormalities. Various metrics of interest such as fetal abdominal circumference (AC) and expected fetal weight may be estimated. However, obtaining the correct plane required for these measurements requires expertise. Such expertise is scarce in low-resource settings where skilled sonographers are not available but have to make do with Community Health Volunteers (CHVs). In such scenarios, AI assistance for standard plane detection and automated measurements of clinically relevant metrics might equip CHVs in screening for early abnormalities. Towards the above goal of providing AI assistance, we will explore various deep-learning-based methods available in the literature to push the state-of-the-art in the above task. References [1] Confident head circumference measurement from ultrasound with real-time feedback for sonographer, Khanal et.al.
https://arxiv.org/pdf/1908.02582 Position 2: Low field Paediatric MRI segmentation and Quality assurance Supervisor: Bishesh Khanal, NAAMII RAs at TOGAI Research Group Summary 3D Imaging modalities such as MRI provide valuable visualization of the developing human brain and help diagnose various developmental disorders. Various publicly available deep-learning-based segmentation models of various organs of interest in Adult MRI are also readily available for such systems to assist in automated quantification. But such devices are not readily available in Low- and Middle-income countries and are associated with high costs of operation and maintenance. On the other hand, low field portable MRI such as Hyperfine SWOOP scanners have shown considerable promise in increasing access and improving healthcare delivery in pediatrics. These low-field scanners can be greatly useful to the underserved community but require Image Quality Assurance tools and specialized Deep-learning based models. In this project, we will explore various techniques for automated quantification of image quality and segmentation of organs of interest. In addition, we will focus on adapting various publicly available adult MRI segmentation models to the pediatric population. References [1] Arnold, Thomas Campbell, et al. “Low‐field MRI: clinical promise and challenges.”
Journal of Magnetic Resonance Imaging 57.1 (2023): 25-44. Position 3: Deep-learning-based Disease Prediction Models for Population Screening Supervisor: Bishesh Khanal, NAAMII RAs at TOGAI Research Group Summary Screening a large population for early diagnosis of diseases is an important public healthcare service. Unfortunately, the scarcity of adequate expertise and equipment hinders widespread screening of key diseases with high burden. To assist in increased screening in low-resource settings, Deep-learning-based methods have shown considerable promise. In this project, we will work with a Local Healthcare Partner towards building the data collection, annotation, training, and evaluation pipeline for building a disease prediction model for screening of the population in a telemedicine setup to provide early diagnosis of diseases for community health centre visitors. Prerequisite Python Programming, some experience with Deep Learning frameworks such as Pytorch or TensorFlow, sound mathematical knowledge, experience in working with medical imaging modalities is a plus, keen interest in Medical Imaging Applications. Guidance and Opportunity for Paper writing The intern will meet with the supervisors for at least half an hour every week and will also be able to interact daily with RAs at NAAMII if needed. There will be close supervision, guidance, training, and monitoring of the project. There will also be an opportunity to write a good scientific paper for the intern, under our guidance if the work done during the internship is deemed suitable. There might also be an opportunity to continue as a research assistant in the future based on mutual interest. Application method Fill this application form with a cover letter and a detailed CV (including any accomplishments or certifications that show your experience in Machine Learning and Computer Vision) as a SINGLE PDF. Plagiarism and Writing Please be aware that plagiarism is a crime, and we have very strong policies against it. When you write any documents, you should not copy even a single sentence from elsewhere! If needed, you can
quote
exactly what has been written elsewhere with due credit (citation). All texts that are
not quoted are expected to be written on your own .
If you are using ideas or information provided by others, you will need to cite the author(s) or sources from where you have borrowed those ideas or information . It is perfectly fine to borrow or bring information from others, but this must be acknowledged with due credit. Writing We understand that good scientific writing requires proper training, practice, and patience. Most students in Nepal are not well trained to do that. Moreover, many are not confident in their English writing skills. We are not going to judge you on how good you write your sentences or how well you write in English. So please do not worry about that. It is much more important to write what you think in your own way even if that is not very well written. This gives us an opportunity to help you learn and improve your scientific writing skills. Duration:
6 months Minimum number of hours per week required:
12 Type of Engagement:
Unpaid Internship Application Deadline:
On a rolling basis till July 15th, 2024
#J-18808-Ljbffr
Are you interested in gaining practical experience in cutting-edge research and technology? Do you have a passion for solving complex problems in deep learning and medical imaging? NAAMII is pleased to announce its Internship & Training Program for the year 2024. This opportunity offers an exciting chance for talented individuals to work on impactful projects under the guidance of experienced supervisors and researchers. Position 1: Operator-agnostic Abdominal Circumference Ultrasound Measurement Supervisor: Bishesh Khanal, Safal Thapaliya Summary Ultrasound is a low-cost medical imaging modality providing excellent visualization of internal body structures without the adverse effects of radiation. Hence, it is a preferred modality for diagnosing pregnancy-related abnormalities. Various metrics of interest such as fetal abdominal circumference (AC) and expected fetal weight may be estimated. However, obtaining the correct plane required for these measurements requires expertise. Such expertise is scarce in low-resource settings where skilled sonographers are not available but have to make do with Community Health Volunteers (CHVs). In such scenarios, AI assistance for standard plane detection and automated measurements of clinically relevant metrics might equip CHVs in screening for early abnormalities. Towards the above goal of providing AI assistance, we will explore various deep-learning-based methods available in the literature to push the state-of-the-art in the above task. References [1] Confident head circumference measurement from ultrasound with real-time feedback for sonographer, Khanal et.al.
https://arxiv.org/pdf/1908.02582 Position 2: Low field Paediatric MRI segmentation and Quality assurance Supervisor: Bishesh Khanal, NAAMII RAs at TOGAI Research Group Summary 3D Imaging modalities such as MRI provide valuable visualization of the developing human brain and help diagnose various developmental disorders. Various publicly available deep-learning-based segmentation models of various organs of interest in Adult MRI are also readily available for such systems to assist in automated quantification. But such devices are not readily available in Low- and Middle-income countries and are associated with high costs of operation and maintenance. On the other hand, low field portable MRI such as Hyperfine SWOOP scanners have shown considerable promise in increasing access and improving healthcare delivery in pediatrics. These low-field scanners can be greatly useful to the underserved community but require Image Quality Assurance tools and specialized Deep-learning based models. In this project, we will explore various techniques for automated quantification of image quality and segmentation of organs of interest. In addition, we will focus on adapting various publicly available adult MRI segmentation models to the pediatric population. References [1] Arnold, Thomas Campbell, et al. “Low‐field MRI: clinical promise and challenges.”
Journal of Magnetic Resonance Imaging 57.1 (2023): 25-44. Position 3: Deep-learning-based Disease Prediction Models for Population Screening Supervisor: Bishesh Khanal, NAAMII RAs at TOGAI Research Group Summary Screening a large population for early diagnosis of diseases is an important public healthcare service. Unfortunately, the scarcity of adequate expertise and equipment hinders widespread screening of key diseases with high burden. To assist in increased screening in low-resource settings, Deep-learning-based methods have shown considerable promise. In this project, we will work with a Local Healthcare Partner towards building the data collection, annotation, training, and evaluation pipeline for building a disease prediction model for screening of the population in a telemedicine setup to provide early diagnosis of diseases for community health centre visitors. Prerequisite Python Programming, some experience with Deep Learning frameworks such as Pytorch or TensorFlow, sound mathematical knowledge, experience in working with medical imaging modalities is a plus, keen interest in Medical Imaging Applications. Guidance and Opportunity for Paper writing The intern will meet with the supervisors for at least half an hour every week and will also be able to interact daily with RAs at NAAMII if needed. There will be close supervision, guidance, training, and monitoring of the project. There will also be an opportunity to write a good scientific paper for the intern, under our guidance if the work done during the internship is deemed suitable. There might also be an opportunity to continue as a research assistant in the future based on mutual interest. Application method Fill this application form with a cover letter and a detailed CV (including any accomplishments or certifications that show your experience in Machine Learning and Computer Vision) as a SINGLE PDF. Plagiarism and Writing Please be aware that plagiarism is a crime, and we have very strong policies against it. When you write any documents, you should not copy even a single sentence from elsewhere! If needed, you can
quote
exactly what has been written elsewhere with due credit (citation). All texts that are
not quoted are expected to be written on your own .
If you are using ideas or information provided by others, you will need to cite the author(s) or sources from where you have borrowed those ideas or information . It is perfectly fine to borrow or bring information from others, but this must be acknowledged with due credit. Writing We understand that good scientific writing requires proper training, practice, and patience. Most students in Nepal are not well trained to do that. Moreover, many are not confident in their English writing skills. We are not going to judge you on how good you write your sentences or how well you write in English. So please do not worry about that. It is much more important to write what you think in your own way even if that is not very well written. This gives us an opportunity to help you learn and improve your scientific writing skills. Duration:
6 months Minimum number of hours per week required:
12 Type of Engagement:
Unpaid Internship Application Deadline:
On a rolling basis till July 15th, 2024
#J-18808-Ljbffr
