Table 2 Types of adoption curves and associated technologies

(I) Continuous increase

\(\frac{\mathrm{d}f(t)}{\mathrm{d}t}=: m\ (t)>0\)

∀  t ∈ [2006; 2017]

HCI, MR-PTC, S-ICD, IAHEI, SCO-MAGN, BS-PV, SE-DES, TAVI, BS-VSAV, DCB-SUAV, DCB-LAV, DCB-AV, SCB-ULV, DCB-LLV, DCB-ARTV (n = 15)

(II) Continuous decrease

\(\frac{\mathrm{d}f(t)}{\mathrm{d}t}=: m\ (t)<0\)

∀  t ∈ [2006; 2017]

SE-BMS (n = 1)

(III) Reaching a saturation

\(\frac{f(2015)}{f(2014)}<1.1\,\wedge\)

\(\frac{f(2016)}{f(2015)}<1.075\,\wedge\)

\(\frac{f(2017)}{f(2016)}<1.05\)

f(2015) > f(t)  ∀   t < 2015

EL-P/ICD, MRD, DCB-CV (n = 3)

(IV) “Local maximum”: Continuous increase followed by continuous decrease

m(t_I) > 0 ∧

m(t_II) < 0

t_I ∈ [t_i; t_j], t_II ∈ [t_j; t_k]

t_i, t_j < t_k

FE-AAA, ACT, ACCS, DEB-TACE, LVRC, BVS, BRA, IABC, MVR, FD-IV (n = 10)

(V) “Local minimum”: Continuous decrease followed by continuous increase

m(t_I) < 0 ∧

m(t_II) > 0

t_I ∈ [t_i; t_j], t_II ∈ [t_j; t_k]

t_i, t_j < t_k

MVAC, FD-ULV (n = 2)

(VI) Complex

NA

 (VI.a)

According to hierarchical-agglomerative clustering method (see Additional file 2)

PECLA/iLA, pVAD, EVCT, CBS, ACD, VEPTR, IAELC, UD-DJMS, DES-LLV, EABO, HCO, IAVC, SP-ENDOST, FDT, DES-SAA, MESI, EBV, NEURO, IABC-EL, DES-ULV, PVR, DCB-TV, DCB-OTHV (n = 23)

 (VI.b)

According to hierarchical-agglomerative clustering method (see Additional file 2)

IET-MICRO, F-TUR, DCB-IV, DCB-VV, ER-ABL (n = 5)