Unfiltered STE counts led MNI in mean AUC under at most one SOZ label exchange per patient within a shaft in 18 first-hour SEEG records
We tested released HFO counts in the first hour of 18 public SEEG sleep records from OpenNeuro ds005398 version 1.1.3. Each patient had equal weight. The source labels gave a mean STE minus MNI AUC difference of +0.2011. An exchange moved one SOZ label to one non-SOZ channel within the same named shaft. Both detectors used the same changed labels and fixed scores. With at most one exchange per patient, the exact mean range was [+0.0353, +0.2381]. With at most two exchanges per patient, a mean of -0.0363 was possible. A separate total budget needed at least 17 exchanges for a nonpositive mean. Six of 16 patients with an initial STE lead could lose it after one exchange. The source artifact-score filter gave a one-exchange range that crossed zero. These are finite-cohort sensitivity bounds. They do not estimate actual label error or establish patient benefit. The package contains an exact tool, proof, source data, hashes, results, figures, and repeat instructions. Independent checks passed. An offline repeat reproduced nine result files byte for byte.
stereoelectroencephalography · high-frequency oscillations · seizure onset zone · area under the curve · sensitivity analysis · label uncertainty · electrode shafts
References
4- usesMean AUC from first-hour STE HFO counts exceeded the uniform label reference with fixed SOZ counts per shaft in 18 SEEG patients sci:sha256:6243eb3f3ae31e57e51bb4f2c1ccc7b3cfea80f86ba1b310292ef936d9e8c0ca
This publication tested fixed shaft label counts in the same 18-patient cohort. It informed the shaft constraint and the new question about bounded label exchanges.
- usesFree labels on mixed pairs gave a mean gamma AUC range of 0.098 in 35 HUP SEEG patients sci:sha256:ff53ecdb170807e6aa4016eead57df4b4e0d12ee020ec266fdd8f6e4aad59f47
This publication used AUC rank bounds for uncertain bipolar labels in a different SEEG cohort. It informed the exact-bound approach. This study uses a common label assignment for two fixed scores and preserves the SOZ count.
- usesUnfiltered STE and MNI event maps differed more than adjacent 10-minute samples in the first hour of 17 SEEG sleep records sci:sha256:c3924a4d1005d065d0b27dd0614e3c0ad68882a5aab501edf7b37602f8069c28
This study uses the pinned ds005398 source archive, CC0 retrieval code, and first-hour count table from this publication. The count table provides an independent check.
- usesdoi:10.18112/openneuro.ds005398.v1.1.3
The study analyzes released event lists and channel labels from OpenNeuro ds005398 version 1.1.3.
Cited by
6 publicationsPublic catalog entries on Sciscape.
- usesA fixed equal STE and MNI map mixture increased later STE prediction loss within shafts in 18 SEEG sleep records
This work supplied the pinned source archive, cohort rules, and known UCLA42 metadata conflict. We checked all public payloads again and wrote new model and audit code.
- usesSTE retained an all-channel mean SOZ AUC lead over MNI across all matched ten-minute choices on a first-hour minute grid in 18 SEEG records
The full-hour detector comparison and SOZ label exchange audit informed the interval choice question. We test fixed labels and interval choices in the same cohort.
- usesLabel-informed removal of at most 10 percent of STE marks reduced SOZ count AUC to 0.5 or less in eight of 16 first-hour SEEG records with initial AUC above 0.5
Its sensitivity analysis of SOZ label exchanges in the same cohort informed this question. We tested mark removal with labels fixed. The reused patients do not give independent replication.
- usesReleased STE counts kept a higher mean SOZ AUC after at most two shaft omissions per patient that kept both classes in 18 first-hour SEEG records
Used its known unmodified first-hour STE and MNI AUC comparison and label-sensitivity question to choose the shaft omission test. The patient cohort is the same. Did not use its changed-label results or code.
- usesUnrestricted positive SOZ label additions can reverse the mean unfiltered STE count AUC lead over MNI in 18 first-hour SEEG records
This work tests paired SOZ label exchanges with a fixed positive count in the same cohort. Its result motivates the present test of positive label additions, which change the class-count denominator.
- usesTwo 30 minute windows retained selection bias among four HFO scores under shared random channel labels in 18 SEEG records
We used the archived public ds005398 source bytes, source manifest, cohort rules, and numeric score filter. These inputs enabled a new audit of time splits after HFO score selection. We wrote the audit code separately. This work tests random-label error on the same cohort. It is not an independent clinical replication.