2026-10-13T13:36:21Z
https://api.figshare.com/v2/oai
oai:figshare.com:article/33868298
2026-09-16T20:20:20Z
category_21
category_106
category_122
portal_5
item_type_3
month_year_09_2026
<p>Subgroup analysis: paired contrasts vs. Text-only on the clinically ambiguous subgroup (short-speech <0.89 s non-Healthy, <i>n</i> = 77 clips from 13 participants). Fold-level paired <i>t</i>-tests on 18 of 25 GroupKFold folds (per the pre-specified small-stra<i>t</i>um exclusion criterion defined in Methods). Only Audio+Text shows a statistically detectable advantage over Text-only at . A sensitivity analysis replacing the 7 excluded folds with and re-testing on all 25 folds yields <i>t</i><sub>24</sub> = 2.26, <i>p</i> = 0.033 for Audio+Text vs. Text-only, confirming the result is not driven by the exclusion. This subgroup contrast is exploratory and uncorrected; see Discussion (<i>Statistical multiplicity and the post-hoc nature of subgroup contrasts</i>) for the multiplicity caveat.</p>
Taiga Noguchi (24999308)
Shotaro Doki (7067663)
Masakazu Hirokawa (24999311)
Soma Nishimura (24999314)
Katsuya Hotta (24999317)
Yuya Iwata (24999320)
Naoko Kouda (24999323)
Shota Matsumoto (6754745)
Kenji Suzuki (157178)
Biotechnology
Science Policy
Mental Health
require psychiatrist interpretation
productivity loss worldwide
prior logit adjustment
mental health problems
learned late fusion
holm – bonferroni
findings provide localised
consensus label served
clinically meaningful support
817 ]), accuracy
10 – 30
001 ), confirming
requires replication ).
29 participants ).
facing smartphone cameras
healthy participants (<
fold repeated groupkfold
exploratory subgroup analysis
div >< p
>< sub xmlns
text outperformed text
f1 =+ 0
repeated smartphone
ece ).
fold clips
brief smartphone
verbal cues
using front
speech responses
report videos
report instruments
regression head
p </
n </
mutually indistinguishable
multimodal machine
multimodal hypothesis
major source
level work
learning model
integrated using
information conditions
independently rated
function scale
function impairment
estimates work
early detection
<p>Subgroup analysis: paired contrasts vs. Text-only on the clinically ambiguous subgroup (short-speech <0.89 s non-Healthy, <i>n</i> = 77 clips from 13 participants). Fold-level paired <i>t</i>-tests on 18 of 25 GroupKFold folds (per the pre-specified small-stra<i>t</i>um exclusion criterion defined in Methods). Only Audio+Text shows a statistically detectable advantage over Text-only at . A sensitivity analysis replacing the 7 excluded folds with and re-testing on all 25 folds yields <i>t</i><sub>24</sub> = 2.26, <i>p</i> = 0.033 for Audio+Text vs. Text-only, confirming the result is not driven by the exclusion. This subgroup contrast is exploratory and uncorrected; see Discussion (<i>Statistical multiplicity and the post-hoc nature of subgroup contrasts</i>) for the multiplicity caveat.</p>
2026-09-16T20:20:02Z
Dataset
Dataset
10.1371/journal.pone.0354342.t003
https://figshare.com/articles/dataset/_p_Subgroup_analysis_paired_contrasts_vs_Text-only_on_the_clinically_ambiguous_subgroup_short-speech_0_89_s_non-Healthy_i_n_i_77_clips_from_13_participants_Fold-level_paired_i_t_i_-tests_on_18_of_25_GroupKFold_folds_per_the_pre-specified_sm/33868298
CC BY 4.0