Setup

using Unfold
using UnfoldMakie, CairoMakie
using UnfoldSim
using UnfoldDecode
using DataFrames
using Statistics

Quick start of Back-To-Back (b2b)

Data generation

Simulation and data collection

Generate single channel data via UnfoldSim.jl

dat, evts = UnfoldSim.predef_eeg(; noiselevel = 0.1, return_epoched = true);

dat is a time x repetition Matrix, evts is a DataFrame`with independent variables / features to explain the data

Dimension expansion

Repeat the dat 20 times, representing 20 channels. In the future we will replace this with a direct multi-channel simulation.

dat_3d = permutedims(repeat(dat, 1, 1, 20), [3 1 2]);

Channels shouldnt be identical, so we add some noise.

dat_3d .+= 0.1 * rand(size(dat_3d)...);

one could specify the specific solvers for G and H by passing the solver_G and solver_H parameters to the solver_b2b function. Implemented solvers are ridge, lasso, lsq, svm, but other solvers from MLJ.jl can be used as well.

Generate the formula

We want to decode condition, but simultaneously control for the effect of continuous.

f = @formula 0 ~ 1 + condition + continuous
time = range(0, 0.44, step = 1 / 100)
designDict = [Any => (f, time)]

m = Unfold.fit(UnfoldModel, designDict, evts, dat_3d; solver = b2b_solver);

results = coeftable(m);
results.estimate = abs.(results.estimate); ## back2back has no sign
results = results[results.coefname.!="(Intercept)", :] ## the intercept in b2b is hard to interpret

90×7 DataFrame

Row	channel	coefname	estimate	eventname	group	stderror	time
	Int64	String	Float64	DataType	Nothing	Nothing	Float64
1	1	condition: face	0.000276096	Any			0.0
2	1	condition: face	0.00121805	Any			0.01
3	1	condition: face	0.00905281	Any			0.02
4	1	condition: face	0.00492913	Any			0.03
5	1	condition: face	0.000306879	Any			0.04
6	1	condition: face	0.0011858	Any			0.05
7	1	condition: face	0.00734451	Any			0.06
8	1	condition: face	0.00833586	Any			0.07
9	1	condition: face	0.00952535	Any			0.08
10	1	condition: face	0.00563764	Any			0.09
11	1	condition: face	0.0590155	Any			0.1
12	1	condition: face	0.187331	Any			0.11
13	1	condition: face	0.576066	Any			0.12
14	1	condition: face	0.304674	Any			0.13
15	1	condition: face	0.269461	Any			0.14
16	1	condition: face	0.27268	Any			0.15
17	1	condition: face	0.27276	Any			0.16
18	1	condition: face	0.286085	Any			0.17
19	1	condition: face	0.310983	Any			0.18
20	1	condition: face	0.389116	Any			0.19
21	1	condition: face	0.0400523	Any			0.2
22	1	condition: face	0.0209925	Any			0.21
23	1	condition: face	0.00728841	Any			0.22
24	1	condition: face	0.00631977	Any			0.23
25	1	condition: face	0.00666009	Any			0.24
26	1	condition: face	0.00641612	Any			0.25
27	1	condition: face	0.00634659	Any			0.26
28	1	condition: face	0.00605966	Any			0.27
29	1	condition: face	0.0064253	Any			0.28
30	1	condition: face	0.00661277	Any			0.29
31	1	condition: face	0.00657537	Any			0.3
32	1	condition: face	0.0062685	Any			0.31
33	1	condition: face	0.00681632	Any			0.32
34	1	condition: face	0.00674751	Any			0.33
35	1	condition: face	0.00723387	Any			0.34
36	1	condition: face	0.00693519	Any			0.35
37	1	condition: face	0.00673979	Any			0.36
38	1	condition: face	0.00701447	Any			0.37
39	1	condition: face	0.00625599	Any			0.38
40	1	condition: face	0.00684818	Any			0.39
41	1	condition: face	0.00893697	Any			0.4
42	1	condition: face	0.00750376	Any			0.41
43	1	condition: face	0.0104718	Any			0.42
44	1	condition: face	0.024929	Any			0.43
45	1	condition: face	0.0225599	Any			0.44
46	1	continuous	0.000147515	Any			0.0
47	1	continuous	6.21685e-5	Any			0.01
48	1	continuous	0.000455905	Any			0.02
49	1	continuous	4.37426e-5	Any			0.03
50	1	continuous	0.000123742	Any			0.04
51	1	continuous	1.94433e-5	Any			0.05
52	1	continuous	8.24115e-5	Any			0.06
53	1	continuous	1.53654e-6	Any			0.07
54	1	continuous	7.52518e-6	Any			0.08
55	1	continuous	0.00014592	Any			0.09
56	1	continuous	8.25061e-7	Any			0.1
57	1	continuous	0.000104262	Any			0.11
58	1	continuous	5.49448e-6	Any			0.12
59	1	continuous	0.0001882	Any			0.13
60	1	continuous	1.16102e-6	Any			0.14
61	1	continuous	6.31581e-5	Any			0.15
62	1	continuous	1.56448e-6	Any			0.16
63	1	continuous	0.000599661	Any			0.17
64	1	continuous	0.00728231	Any			0.18
65	1	continuous	0.0978105	Any			0.19
66	1	continuous	0.854214	Any			0.2
67	1	continuous	0.355671	Any			0.21
68	1	continuous	0.282977	Any			0.22
69	1	continuous	0.285093	Any			0.23
70	1	continuous	0.278331	Any			0.24
71	1	continuous	0.282962	Any			0.25
72	1	continuous	0.284605	Any			0.26
73	1	continuous	0.284962	Any			0.27
74	1	continuous	0.283158	Any			0.28
75	1	continuous	0.281591	Any			0.29
76	1	continuous	0.284366	Any			0.3
77	1	continuous	0.282344	Any			0.31
78	1	continuous	0.284764	Any			0.32
79	1	continuous	0.285409	Any			0.33
80	1	continuous	0.281967	Any			0.34
81	1	continuous	0.280067	Any			0.35
82	1	continuous	0.281299	Any			0.36
83	1	continuous	0.274928	Any			0.37
84	1	continuous	0.278807	Any			0.38
85	1	continuous	0.267941	Any			0.39
86	1	continuous	0.259431	Any			0.4
87	1	continuous	0.239387	Any			0.41
88	1	continuous	0.1926	Any			0.42
89	1	continuous	0.0523001	Any			0.43
90	1	continuous	0.000115614	Any			0.44

Plotting

plot_erp(results; axis = (xlabel = "Time [s]", ylabel = "Performance"))

We can see from the graph that b2b solver identifies regions where the signal can be decoded, taking into account the continuous feature

This page was generated using Literate.jl.

Setup

Quick start of Back-To-Back (b2b)

Data generation

Simulation and data collection

Dimension expansion

Modeling

Solver selection

Generate the formula

Plotting