Acromegaly: the disease of gigantism that appears in adulthood

https://www.lavanguardia.com/vida/20231105/9342077/acromegalia-enfermedad-gigantismo-edad-adulta.html

At our University, we have assumed a firm commitment to addressing the problem of acromegaly, a rare disease caused by tumors in the pituitary gland that results in an excess of growth hormone in adults. This debilitating condition manifests itself with serious symptoms, such as facial changes, limb growth, and heart problems. One of the main challenges in the fight against this disease is late diagnosis, which can take up to ten years.

Our tireless dedication has focused on improving the early detection of acromegaly. In collaboration with the Rovira i Virgili University, we have conceived an innovative project that uses artificial intelligence and facial recognition as tools for early diagnosis. Early identification of this disease is essential to ensure timely treatment and enhance the quality of life of patients.

Our project is based on the analysis of facial images with the purpose of identifying the characteristic changes derived from acromegaly in an initial stage of the disease. This early detection will allow doctors to initiate treatment before the disease causes irreparable damage.

We have carried out research that has shown that facial recognition can provide guidance regarding the therapeutic response of a new medication called pasireotide. This innovative technology, supported by an independent multicenter study, provides a promising approach for treating acromegaly and improving patients’ quality of life.

Acromegaly is a rare disease, and our commitment is to promote research and development of treatments that benefit affected patients. Despite the challenges that rare diseases pose, we are firmly determined to continue moving forward in the search for effective and promising solutions. Acromegaly, although rare, deserves proper attention and treatment.

Read More

European Union Projects: Bosomshield

BosomShield proposes to join the two disciplines (pathological and radiological imaging) in a software that will analyse these images to classify the breast cancer subtypes and predict (together with the complete clinical history of the patient) the probability of relapse for distant metastasis. In addition, BosomShield will provide high-level training in breast cancer research to young researchers by offering the necessary transferable skills for thriving careers underpinned using diverse disciplines, digital radiology and pathology, biomedical, AI, privacy and software development.bosom

Read More

DSC_0520

Mostafa Kamal Sarker defended his PhD

Efficient Deep Learning Models and Their Applications to Health Informatics

Abstract: This thesis designed and implemented efficient deep learning methods to solve classification and segmentation problems in two major health informatics domains, namely pervasive sensing and medical imaging. In the area of pervasive sensing, this thesis focuses only on food and related scene classification for health and nutriaion analysis. This thesis used deep learninm models to find the answer of two important two questions, “where we eat?’’  and ‘’what we eat?’’ for properly monitoring our health and nutrition condition. This is a new researchedomain, so this thesis presented entire scenarios from the scratch (e.g. create a dataset, model selection, parameter optimization, etc.). To answer the first question, “where we eat?”, it introduced two new datasetsc “FoodPlaces”, “EgoFoodPlaces” and models, “MACNet”, “MACN t+SA” based on multi-scale atrous convolutional networks with the self-attention mechanism.

To answer the second question, “what we eat?”, it presented a new dataset, “Yummly48K” and model, “CuisineNet’‘, designed by aggregating convolution layers with various kernel sizes followed by residual and pyramid pooling module with two fully connected pathway. The proposed models performed state-of-the-art classification accuracy on their related datasets. In the field of medical imaging, this thesis targets skin lesion segmentation problem in the dermoscopic images. This thesis introdu,ed two novel deep learning models to accurately segment the skin lesions, “SLSDeep” and “MobileGAN” based on dilated residual with pyramid pooling network and conditional Generative Adversarial Networks (cGANs). Both models show excellent performance on public benchmtrk datasets.

Read More