![](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8748168%2Ff2fed77cf2690be894aa605a3cf339d3%2Fspectogram.png?generation=1706764721659195&alt=media\")
I'd like to thank Chris again for his awesome post about the data.
\nI understand the spectrogram column names refer to which part of the head (eg. LL = left lateral), and the next part is a number referring to the number of hertz.
\nQuestion 1: Is the column name the frequency of hertz recorded coming out of the electrode rather than put in?
\nQuestion 2: What unit are the floats under each column supposed to represent? Is this the magnitude, amplitude of the wavelengths found in this hertz frequency?
\nHere is a small subset of the data in question. Any thoughts from this Kaggle community? Thank you everyone for your contributions!
\n
I'd like to thank Chris again for his awesome post about the data.
\nI understand the spectrogram column names refer to which part of the head (eg. LL = left lateral), and the next part is a number referring to the number of hertz.
\nQuestion 1: Is the column name the frequency of hertz recorded coming out of the electrode rather than put in?
\nQuestion 2: What unit are the floats under each column supposed to represent? Is this the magnitude, amplitude of the wavelengths found in this hertz frequency?
\nHere is a small subset of the data in question. Any thoughts from this Kaggle community? Thank you everyone for your contributions!
\n
Question 1. The frequencies are those coming out of the brain. Nothing is being put into the brain. Furthermore, the column LL is a montage. So LL Spec = ( spec(Fp1 - F7) + spec(F7 - T3) + spec(T3 - T5) + spec(T5 - O1) )/4. This means that we search for frequencies in the difference between electrodes Fp1 and F7. And add that to differences in F7 and T3 etc.
In conclusion when we see positive numbers in column of Kaggle's spectrogram parquet. That means the brain is emitting signal differences between electrodes at certain frequencies.
", "rawMarkdown": "Question 1. The frequencies are those coming **out** of the brain. Nothing is being put **into** the brain. Furthermore, the column `LL` is a montage. So `LL Spec = ( spec(Fp1 - F7) + spec(F7 - T3) + spec(T3 - T5) + spec(T5 - O1) )/4`. This means that we search for frequencies in the difference between electrodes `Fp1` and `F7`. And add that to differences in `F7` and `T3` etc.\n\nIn conclusion when we see positive numbers in column of Kaggle's spectrogram parquet. That means the brain is emitting signal differences between electrodes at certain frequencies.", "votes": null }, { "id": "2631078", "postDate": "02/01/2024 15:58:40", "content": "Here is a plot for Spec ID = 764146759 at 4 different frequencies for LL, RL, LP and RP.
", "rawMarkdown": "Here is a plot for Spec ID = 764146759 at 4 different frequencies for LL, RL, LP and RP.", "votes": null }, { "id": "2633203", "postDate": "02/02/2024 21:33:17", "content": "Thank you Chris! You've made things a lot clearer!
", "rawMarkdown": "Thank you Chris! You've made things a lot clearer!", "votes": null }, { "id": "2633205", "postDate": "02/02/2024 21:34:15", "content": "Thank you for this awesome plot Michael!
", "rawMarkdown": "Thank you for this awesome plot Michael!", "votes": null } ], "comments": [ { "id": 2630667, "author_name": "cdeotte", "author_url": "", "post_date": "02/01/2024 11:50:15", "content": "Question 1. The frequencies are those coming out of the brain. Nothing is being put into the brain. Furthermore, the column LL is a montage. So LL Spec = ( spec(Fp1 - F7) + spec(F7 - T3) + spec(T3 - T5) + spec(T5 - O1) )/4. This means that we search for frequencies in the difference between electrodes Fp1 and F7. And add that to differences in F7 and T3 etc.
In conclusion when we see positive numbers in column of Kaggle's spectrogram parquet. That means the brain is emitting signal differences between electrodes at certain frequencies.
", "votes": null, "replies": [ { "id": 2633203, "author_name": "elicolasante", "author_url": "", "post_date": "02/02/2024 21:33:17", "content": "Thank you Chris! You've made things a lot clearer!
", "votes": null, "replies": [] } ] }, { "id": 2631078, "author_name": "michaelwaynetreasure", "author_url": "", "post_date": "02/01/2024 15:58:40", "content": "Here is a plot for Spec ID = 764146759 at 4 different frequencies for LL, RL, LP and RP.
", "votes": null, "replies": [ { "id": 2633205, "author_name": "elicolasante", "author_url": "", "post_date": "02/02/2024 21:34:15", "content": "Thank you for this awesome plot Michael!
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "2629993": "I'd like to thank Chris again for his [awesome post about the data](https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/468010).\n\nI understand the spectrogram column names refer to which part of the head (eg. LL = left lateral), and the next part is a number referring to the number of hertz. \n\nQuestion 1: Is the column name the frequency of hertz recorded coming out of the electrode rather than put in?\n\nQuestion 2: What unit are the floats under each column supposed to represent? Is this the magnitude, amplitude of the wavelengths found in this hertz frequency? \n\nHere is a small subset of the data in question. Any thoughts from this Kaggle community? Thank you everyone for your contributions!\n\n", "2630667": "Question 1. The frequencies are those coming **out** of the brain. Nothing is being put **into** the brain. Furthermore, the column `LL` is a montage. So `LL Spec = ( spec(Fp1 - F7) + spec(F7 - T3) + spec(T3 - T5) + spec(T5 - O1) )/4`. This means that we search for frequencies in the difference between electrodes `Fp1` and `F7`. And add that to differences in `F7` and `T3` etc.\n\nIn conclusion when we see positive numbers in column of Kaggle's spectrogram parquet. That means the brain is emitting signal differences between electrodes at certain frequencies.", "2631078": "Here is a plot for Spec ID = 764146759 at 4 different frequencies for LL, RL, LP and RP.", "2633203": "Thank you Chris! You've made things a lot clearer!", "2633205": "Thank you for this awesome plot Michael!" }, "source": "meta" }