![]() ![]() The CNN technique developed here may offer enhanced utility in the design and development of head protective countermeasures, including in the automotive industry. Therefore, the CNN has the potential to supersede current kinematic injury metrics that can only approximate a global peak MPS or CSDM. Importantly, the CNN is able to efficiently estimate elementwise MPS with sufficient accuracy while conventional kinematic injury metrics cannot. Finally, the CNN achieved an average k and r of 0.98☐.12 and 0.90☐.07, respectively, for six reconstructed car crash impacts drawn from two other sources independent of the training dataset. Cumulative strain damage measure (CSDM) from the CNN estimation was also highly accurate compared to those from direct simulation across a range of thresholds (R2 of 0.899–0.943 with RMSE of 0.054–0.069). It also achieved a success rate of 60.5% and 94.8% for elementwise MPS, where the linear regression slope, k, and correlation coefficient, r, between estimated and simulated MPS did not deviate from 1.0 (when identical) by more than 0.1 and 0.2, respectively. For peak MPS, the CNN achieved a coefficient of determination (R2) of 0.932 and root mean squared error (RMSE) of 0.031 for the real-world testing dataset. The combined training achieved the best performances. Three training strategies were evaluated: 1) “baseline”, using random initial weights 2) “transfer learning”, using weight transfer from a previous CNN model trained on head impacts drawn from contact sports and 3) “combined training”, combining previous training data from contact sports (N=5661) for training. For each augmented impact, rotational velocity (v_rot) and the corresponding rotational acceleration (a_rot) profiles were concatenated as static images to serve as CNN input. They were simulated using the anisotropic Worcester Head Injury Model (WHIM) V1.0, which provided baseline elementwise peak maximum principal strain (MPS). Head impact kinematics (N=458) from two public databases were used to generate augmented impacts (N=2694). ![]() Here, we extend its application to automotive head impacts, where impact profiles are typically more complex with longer durations. Lately, a convolutional neural network (CNN) has been successfully developed to estimate spatially detailed brain strains instantly and accurately in contact sports. More from our Marijuana archive: " Medical marijuana software: Developers MJ Freeway and idWeeds roll out wired weed.Efficient brain strain estimation is critical for routine application of a head injury model. However, we're not expecting these weed whizzes to get by the Apple Store censors without a hell of a fight. ![]() That's coming later this year, according to the company. Plus, Strain Brain hasn't yet rolled out the smart phone apps it promises will offer the same function. It's the sort of tech that could prove useful in cutting through the confusion and false promises of the current MMJ market - that is, if it works. The tech involved is presumably similar to facial-recognition technology, although we prefer to imagine a sweatshop filled with stoners chained to computer monitors somewhere in Silicon Valley who have to identify A-Train and Purple Haze for hours on end, day and night. All you have to do is upload an image of an indeterminate strain, and, according to the company's website, "top secret space-age strain recognition technology will quickly identify the exact strain of marijuana you're smoking," as well as locate nearby medical marijuana centers offering the particular meds. Strain Brain's website is already up and running. If Strain Brain makes good on its word, they will soon be able to - courtesy of the company's strain-recognition technology. But can they help you tell a sample of Green Crack from G-13 Haze? Smart phones can already do all sorts of clever things, like tell you where the nearest medical marijuana center is located.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |