These mask-induced OSD issues that doctors are seeing in their chairs are mirrored in the literature. A recent study that looked at the effects of daily face mask wear on the ocular surface found that improper use may lead to increased ocular irritation and DED symptoms.1 Additionally, a case report published in Eye & Contact Lens found a heightened number of DED symptoms in regular mask-wearing patients with no previous history of ocular surface issues.2
Dr. Brujic cites an upward trend of patients presenting with internal and external hordeolum caused by mask wear. In addition to styes, optometrist Scott Hauswirth, assistant professor at the University of Colorado School of Medicine, is also encountering more cases of blepharitis, which he attributes to improper mask wear.
The Beneath The Mask
Specifically, poor face mask fits result in increased airflow over the ocular surface, which in turn increases the evaporation of the tear film and compromises its protective barrier, resulting in dry eyes.2 In individuals who already have a compromised epithelial surface, this can increase their susceptibility to infection.2
In addition to bacterial infiltration, Dr. Karpecki cites another reason for OSD-related issues caused by mask wear: the desiccation of the ocular surface, which he says is a similar process to what can happen to the eye during continuous positive airway pressure (CPAP) overnight treatments.
Another investigation found long-term use of face masks not only shortened TBUT time but also reduced Schirmer-1 measurements.1 The research paper cited a significant increase in OSDI scores and ocular surface staining according to the Oxford scale after eight hours of continuous face mask use.1 Furthermore, it was observed that taping down the open upper portion of the mask improved test parameters by preventing exhaled dry air from contacting the eyes.1
Dr. Karpecki is seeing more patients with inadequate lid closures with these mask-related OSD issues. This condition should not be confused with lagophthalmos, the inability to close the eyelids, he suggests.
Additionally, some patients are bringing up their masks too high and limiting their lower eyelid from fully making contact with the upper eyelid, so placement or positioning of the mask is important, Dr. Silani adds.
Researchers from Wills Eye Hospital who looked into bacterial dispersion with patient face mask use during simulated intravitreal injections found alterations of bacterial dispersion around the eye during treatment based on the type of mask worn.10 The authors noted significantly more bacterial dispersion when wearing a tight-fitting face mask without tape vs. a tight-fitting one with tape. Additionally, their study found no difference in bacterial dispersion between tight-fitting surgical masks with tape and N95 masks.8
In addition to recommending patients tape their mask on the bridge of the nose to minimize an upward draft toward the eye and take care not to affect the normal blink dynamics or eyelid closure, Dr. Shovlin also suggests reminding the patient to avoid touching their face or adjusting their mask as much as possible.
Lid margin integrity and meibum function improving due to office-based treatments in a 27-year-old female patient with DED secondary to MGD. She works in a healthcare facility and prolonged face mask wear has exacerbated her dry eye condition. Photo: Kambiz Silani, OD. Click image to enlarge.
Exploring the impact of contact lenses and mask wear, a team of researchers from Portugal found mask use increased dry eye symptoms in contact lens wearers and negatively impacted their visual quality.11 Additionally, lens wear significantly decreased after the onset of COVID, and the sensation of dry eye was found to be worse in those using monthly replacement contact lenses.11
With face mask use likely extending into at least the near future, eye dryness, irritation and keratopathy from mask wear may become a problem for a large percentage of the population.14 If present, this mask-associated ocular irritation raises concerns about eye health and increased risk of disease transmission with prolonged wear.12
Considering this forecast, additional investigations may be warranted. One consideration for future studies could be an analysis of geographical differences from one region to another and a comparison of ocular surface prevalence issues in areas that are more compliant in mask-wearing practices or in areas with higher rates of COVID-19, Dr. Shovlin suggests.
Along with mask-induced dry eye, Dr. Silani also stresses that it is important to consider potential comorbidities, such as lagophthalmos, ocular allergies and anterior blepharitis, as multiple, contributing factors could be at play.
Virtually everyone has to wear a public mask in order to be accepted by others. That's a simple fact of human psychology. When circumstances (such as anonymity, strong emotion, or sufficient power) allow a character to take off that mask and act in complete accord with their inclinations, they reveal what's beneath the mask.
Sometimes a person may never know they had a hidden self before the mask comes off. The change is even a surprise to them. Other times the person is well aware of their hidden self and are determined to keep it hidden. This hidden self that people don't show to others is what Beneath the Mask is about.
The concept in Western philosophy originated with Carl Jung, who referred to the mask people wear in public as the "persona" (Latin via Ancient Greek for "mask"), and their hidden desires as their "shadow."
As players that peruse this list are sure to notice, the numbers 9, 40, and 49 are repeated several times across the What Lies Beneath the Mask Special Research tasks and rewards. For those fans that are curious about the significance of those numbers, they are references to evolving Galarian Yamask in Pokemon Sword and Shield. More specifically, Yamask must take 49 damage before it will evolve into Runerigus in those games.
After the exhibition, Lucretia prepares to move onto the next event of the evening when Tullius interrupts with his sudden arrival. Despite this, Lucretia continues, afraid to force him to leave. Painted slaves, both male and female, are brought forth for the guests. Tullius decides that, before donning a mask, he should put himself to a challenge and fight Gannicus. Oenomaus tells Gannicus to indulge Tullius and lose so that the man is not insulted. Tullius, though skilled with a sword, is no match for Gannicus; even so, the latter allows Tullius to make a fool of him, and gives missio after receiving multiple cuts to the torso.
Masks have become a point of immense contention since the pandemic began, imbued with a symbolic, cultural, political meaning that outstrips any question of their effectiveness. This is unfortunate, because the greater the status of masks as a totem, the less anyone wants to discuss their usefulness as an intervention against the coronavirus. A cost-benefit analysis is out of the question; indeed, people tend to react with derision and hostility if you suggest that there might be costs at all.
We start with a dataset of faces with precomputed face landmarks that are processed through the mask generator. It uses the landmarks to position the mask on the face. Now that we have the dataset with pairs of images (with and without the mask), we can continue with defining the architecture of the ML model. Finally, the last part of the pipeline is to find the best loss function and all necessary scripts to put everything together, so we can train and evaluate the model.
To train the deep learning model, we need a lot of data. In this case, we need a lot of pairs of input and output images. It is of course not practical to gather both input and output images of the same person with and without a mask. We can do better and act much faster.
Next, we need to locate facial landmarks so we can place the mask in the right place. To do so, we used a pretrained dlib facial landmark detector. You can use any other similar dataset, just make sure that you can find the precomputed face landmark points (this GitHub repo can be a good source) or compute landmarks by yourself.
Initially, we started with a simple implementation of a mask generator by placing a polygon on the face with a randomized distance of the polygon vertices from the face landmarks. This way, we could quickly generate a simple dataset and test if the idea behind this project is valid. Once established that it really is valid, we looked for a more robust solution that would better capture real-world scenarios.
There is a great GitHub project, Mask The Face, that already solves the mask generation problem. It estimates mask position from face landmark points, estimates face tilt angle to select the best fitting mask from the database and, finally, places the mask on the face. The database of available masks includes surgical masks, cloth masks with a variety of colors and textures, several types of respirators and even gas masks.
First, the skip connections are known to speed up the learning process and help tackle the vanishing gradient problem [5]. Second, they pass the information from the encoder directly to the decoder, helping recover information loss during downsampling. We can say that they can propagate all parts of the image outside of the face mask that we want to keep unchanged, while also helping with the generation of the part of the face beneath the mask.
This is exactly what we need! Skipping connections will help preserve the parts of the input that we want to propagate to the output, while the Encoder-Decoder part of the U-net will detect the mask and replace it with the mouth underneath.
Both MSE and MAE are loss functions based on the per-pixel difference between the image generated by our model and the ground truth image from the dataset before applying the mask onto the face. This seems like exactly what we need, but we are not trying to train the model that can pixel-perfectly recreate what is hidden under the mask. 2ff7e9595c
Comments