Eliminating the need for manual segmentation to determine size and volume from MRI. A proof of concept on segmenting the lateral ventricles.

Manual segmentation, which is tedious, time-consuming, and operator-dependent, is currently used as the gold standard to validate automatic and semiautomatic methods that quantify geometries from 2D and 3D MR images. This study examines the accuracy of manual segmentation and generalizes a strategy to eliminate its use. Trained individuals manually measured MR lateral ventricles images of normal and hydrocephalus infants from 1 month to 9.5 years of age. We created 3D-printed models of the lateral ventricles from the MRI studies and accurately estimated their volume by water displacement. MRI phantoms were made from the 3D models and images obtained. Using a previously developed artificial intelligence (AI) algorithm that employs four features extracted from the images, we estimated the ventricular volume of the phantom images. The algorithm was certified when discrepancies between the volumes-gold standards-yielded by the water displacement device and those measured by the automation were smaller than 2%. Then, we compared volumes after manual segmentation with those obtained with the certified automation. As determined by manual segmentation, lateral ventricular volume yielded an inter and intra-operator variation up to 50% and 48%, respectively, while manually segmenting saggital images generated errors up to 71%. These errors were determined by direct comparisons with the volumes yielded by the certified automation. The errors induced by manual segmentation are large enough to adversely affect decisions that may lead to less-than-optimal treatment; therefore, we suggest avoiding manual segmentation whenever possible.

Accurate image-based CSF volume calculation of the lateral ventricles.

The size/volume of the brain's ventricles is essential in diagnosing and treating many neurological disorders, with various forms of hydrocephalus being some of the most common. Initial ventricular size and changes, if any, in response to disease progression or therapeutic intervention are monitored by serial imaging methods. Significant variance in ventricular size is readily noted, but small incremental changes can be challenging to appreciate. We have previously reported using artificial intelligence to determine ventricular volume. The values obtained were compared with those calculated using the inaccurate manual segmentation as the "gold standard". This document introduces a strategy to measure ventricular volumes where manual segmentation is not employed to validate the estimations. Instead, we created 3D printed models that mimic the lateral ventricles and measured those 3D models' volume with a tuned water displacement device. The 3D models are placed in a gel and taken to the magnetic resonance scanner. Images extracted from the phantoms are fed to an artificial intelligence-based algorithm. The volumes yielded by the automation must equal those yielded by water displacement to assert validation. Then, we provide certified volumes for subjects in the age range (1-114) months old and two hydrocephalus patients.

Turning any bed into an intensive care unit with the Internet of things and artificial intelligence technology. Presenting the enhanced mechanical ventilator.

The recent Coronavirus disease 2019 (COVID-19) pandemic displayed weaknesses in the healthcare infrastructures worldwide and exposed a lack of specialized personnel to cover the demands of a massive calamity. We have developed a portable ventilator that uses real-time vitals read from the patient to estimate -- through artificial intelligence -- the optimal operation point. The ventilator has redundant telecommunication capabilities; therefore, the remote assistance model can protect specialists and relatives from highly contagious agents. Additionally, we have designed a system that automatically publishes information in a proprietary cloud centralizer to keep physicians and relatives informed. The system was tested in a residential last-mile connection, and transaction times below the second were registered. The timing scheme allows us to operate up to 200 devices concurrently on these lowest-specification transmission control protocol/internet protocol (TCP/IP) services, promptly transmitting data for online processing and reporting. The ventilator is a proof of concept of automation that has behavioral and cognitive inputs to cheaply, yet reliably, extend the installed capacity of the healthcare systems and multiply the response of the skilled medical personnel to cover high-demanding scenarios and improve service quality.

Measuring Maximum Head Circumference Within the Picture Archiving and Communication System: A Fully Automatic Approach.

This study describes an automatic technique to accurately determine the maximum head circumference (MHC) measurement from MRI studies within the Picture Archiving and Communications System, and can automatically add this measurement to the final radiology report. Participants were selected through a retrospective chart review of patients referred to the neurosurgery clinic. Forty-nine pediatric patients with ages ranging from 5 months to 11 years were included in the study. We created 14 printed ring structures to mirror the head circumference values at various ages along the x-axis of the Nellhaus chart. The 3D-printed structures were used to create MRI phantoms. Analytical obtainment of circumference values from the 3D objects and phantom images allowed for a fair estimation and correction of errors on the image-based-measuring instrument. Then, standard manual MHC measurements were performed and compared to values obtained from the patients' MRI T1 images using the tuned instrument proposed in this document. A T-test revealed no statistical difference between the manual assessments and the ones obtained by the automation p = 0.357, α = 0.05. This automatic application augments the more error-prone manual MHC measurement, and can add a numerical value to the final radiology report as a standard application.

Automatically measuring brain ventricular volume within PACS using artificial intelligence.

The picture archiving and communications system (PACS) is currently the standard platform to manage medical images but lacks analytical capabilities. Staying within PACS, the authors have developed an automatic method to retrieve the medical data and access it at a voxel level, decrypted and uncompressed that allows analytical capabilities while not perturbing the system's daily operation. Additionally, the strategy is secure and vendor independent. Cerebral ventricular volume is important for the diagnosis and treatment of many neurological disorders. A significant change in ventricular volume is readily recognized, but subtle changes, especially over longer periods of time, may be difficult to discern. Clinical imaging protocols and parameters are often varied making it difficult to use a general solution with standard segmentation techniques. Presented is a segmentation strategy based on an algorithm that uses four features extracted from the medical images to create a statistical estimator capable of determining ventricular volume. When compared with manual segmentations, the correlation was 94% and holds promise for even better accuracy by incorporating the unlimited data available. The volume of any segmentable structure can be accurately determined utilizing the machine learning strategy presented and runs fully automatically within the PACS.

Tractography in the clinics: Implementing a pipeline to characterize early brain development.

In imaging studies of neonates, particularly in the clinical setting, diffusion tensor imaging-based tractography is typically unreliable due to the use of fast acquisition protocols that yield low resolution and signal-to-noise ratio (SNR). These image acquisition protocols are implemented with the aim of reducing motion artifacts that may be produced by the movement of the neonate's head during the scanning session. Furthermore, axons are not yet fully myelinated in these subjects. As a result, the water molecules' movements are not as constrained as in older brains, making it even harder to define structure using diffusion profiles. Here, we introduce a post-processing method that overcomes the difficulties described above, allowing the determination of reliable tracts in newborns. We tested our method using neonatal data and successfully extracted some of the limbic, association and commissural fibers, all of which are typically difficult to obtain by direct tractography. Geometrical and diffusion based features of the tracts are then utilized to compare premature babies to term babies. Our results quantify the maturation of white matter fiber tracts in neonates.

Extending PACS Functionality: Towards Facilitating the Conversion of Clinical Necessities into Research-Derived Applications.

The Picture Administration and Communications System (PACS) was designed to replace the old film archiving system in hospitals in order to store and move varying medical image modalities. Using the standard Internet transport protocol, PACS creators designed a robust digital signaling platform to optimize media use, availability, and confidentiality. Nowadays PACS has become ubiquitous in medical facilities but lacks imaging analytical capabilities. A myriad of initiatives have been launched in the hope of achieving this goal, but current solutions face issues with security and ease-of-use that have precluded their widespread adoption. Here, we present a PACS-based image processing tool that safeguards patient confidentiality, is user-friendly and is easy to implement. The final product is platform-independent, has a small degree of intrusiveness and is well suited to clinical and research workflows.

Magnetic resonance image enhancement by reducing receptors' effective size and enabling multiple channel acquisition.

Magnetic resonance imaging is empowered by parallel reading, which reduces acquisition time dramatically. The time saved by parallelization can be used to increase image quality or to enable specialized scanning protocols in clinical and research environments. In small animals, the sizing constraints render the use of multi-channeled approaches even more necessary, as they help to improve the typically low spatial resolution and lesser signal-to-noise ratio; however, the use of multiple channels also generates mutual induction (MI) effects that impairs imaging creation. Here, we created coils and used the shared capacitor technique to diminish first degree MI effects and pre-amplifiers to deal with higher order MI-related image deterioration. The constructed devices are tested by imaging phantoms that contain identical solutions; thus, creating the conditions for several statistical comparisons. We confirm that the shared capacitor strategy can recover the receptor capacity in compounded coils when working at the dimensions imposed by small animal imaging. Additionally, we demonstrate that the use of pre-amplifiers does not significantly reduce the quality of the images. Moreover, in light of our results, the two MI-avoiding techniques can be used together, therefore establishing the practical feasibility of flexible array coils populated with multiple loops for small animal imaging.