Alessandro Ferrari’s Post

Image classification machine learning models . Object detection machine learning models . ( x-ray imaging ) #ai #machinelearning #healthcare #aiinmedicine #micorsoft challenge reflections 🧠

YOWZA ! __ A single 92-plex panel pulls out prediction of death! With all this fear of AI and machines taking over, this 2021 study used AI/machine learning statistical analysis for the betterment of mankind. The only thing better than screening a large cohort (2000 patients) is validating the findings in a second large cohort (800 patients). Predictive values improved from an AUC of 0.67 to 0.94 when a multiplex signature was used over conventional risk scoring. Can you imagine the predictive power if the current full library were screened? (5,400 targets vs 92) "Proteomics-Enabled Deep Learning Machine Algorithms Can Enhance Prediction of Mortality" https://lnkd.in/e38mVcXW (JACC, Baby ! :) Cohorts__ LIFE heart study, PLIC study Markers "discovered"__BNP, CTRC, KIM1, MMP12, TRAILR2 #proteomics, #predictiveanalytics, #heartdisease

Excited to share the publication of two research papers in Sensor Journal: 1. "Gait Event Detection and Travel Distance Using Waist-Worn Accelerometers across a Range of Speeds: Automated Approach" - [DOI link](https://lnkd.in/gjVGEWCm) 2. "Gait Characterization in Duchenne Muscular Dystrophy (DMD) Using a Single-Sensor Accelerometer: Classical Machine Learning and Deep Learning Approaches" - [DOI link](https://lnkd.in/gtrgKtgs) Delighted to contribute to advancements in sensor technology and gait analysis. #DMD #Sensor #GaitAnalysis #AI #DL #ML

🔬 Unlocking the Power of AI-based Image Analysis — Part III 🔬   The last two episodes demonstrated AI's ability to handle complex segmentation tasks, both in fluorescence and color images. However, existing AI models are sometimes limited in their ability to handle unforeseen variations (outlier, poor image quality, change in imaging setup). Fortunately, there's a solution!   The transfer learning 🖥️   In fact, it is possible to manually add relevant labelled data to the database to re-train a model and help it improve performance. At Imactiv-3D, we have developed an open-source segmentation Napari plugin called Sketchpose, enabling us to offer transfer learning on any type of data to improve results and seamlessly achieve better segmentation.   Bacteria counting 🦠   In today’s illustration (1), the image is a phase contrast microscope acquisition of bacteria. Segmentation of contiguous filiform objects such as bacteria is often difficult, and most existing models do not directly lead to perfect segmentation results that would enable reliable characterization. Next is a first unsatisfactory segmentation result with an existing model (2) then improved using our transfer learning method (3), leading to a much better output.   Stay tuned! 📢   Stay tuned for further insights into how our expertise in various imaging techniques can improve scientific analysis.   #Imactiv3D #ImageAnalysis #Segmentation #Research #AI #Napari

Meta made an announcement that they can now create real-time images from brain activity...how cool and scary is that at the same time. Meta is using magnetoencephalography (MEG), a non-invasive neuroimaging technique in which thousands of brain activity measurements are taken per second for this advancement in image processing. You can read more about MEG here https://lnkd.in/dJFQKhB2 #artificialintelligence #ai #generativeai #imageprocessing https://lnkd.in/d3kvVTsN