O painel foi considerado, por vezes, um original do séc. XV com possíveis alterações antigas, outras vezes como uma cópia mais tardia datando do séc. XVI, devido ao estilo da inscrição. O presente consenso aponta na direcção de um original de Rogier Van der Weyden e/ou sua oficina, e a identificação da retratada pode ser considerada provada para além de qualquer dúvida razoável.

A datação dendrocronológica realizada em 2000 e a associação da madeira à de outras obras atribuídas à oficina de van der Weyden, indicaram que a antiguidade da pintura era claramente maior do que a hipótese de cópia tardia pressupunha, fazendo recuar a sua execução inicial para cerca de 1450 e admitindo possiveis alterações importantes de origem. Do relatório do Museu J. Paul Getty datado 2/12/2001 (accession number 78.PB.3), transcrevemos a seguinte passagem:

"Dendrochronological analysis by Dr. Peter Klein, Ordinariat für Holzbiologie, University of Hamburg, September 11, 2000: the two oak boards from the Baltic/Polish region were felled in 1435 at the earliest, with creation of the painting possible from 1437 and plausible from 1451 upwards. Board I is made from the same tree as BI in St. Ivo, and BI in Pieta, National Gallery London, and St. Mary and St. Catherine, Kunsthistorisches Museum, Vienna. See report of October 6, 2000."

Para além do recuo da datação, os estudos científicos realizados detectaram uma surpreendente versão subjacente da pintura em que a duquesa envergava um vestido verde com gola vermelha, anterior à repintura com o vestido de brocados dourados, provavelmente realizada pouco depois, seguindo de perto os contornos das vestes iniciais, e que corresponde ao aspecto actual da pintura, tendo a inscrição sido acrescentada mais tarde. Transcrevemos de Jochen Sander em The Master of Flémalle and Rogier van der Weyden (2009, pág. 364):

"In 2004, now joined by Yvonne Szafran, [Lorne] Campbell studied the painting anew and, thanks primarily to a thorough technological examination, made surprising new findings which strongly suggest that the portrait was painted in Rogier's workshop in Brussels in about 1450. The results of the investigation show that the dress and veil were overpainted relatively soon after the work's initial execution. The duchess was originally depicted wearing a green gown with a red collar and a white belt. Lorne Cambpell proposed the early sixteenth century as the date for this overpainting, a conjecture that remains uncertain, however, in view of the painterly execution. For, as Campbell himself stressed, from the technical point of view, the execution of the brocade still exhibits all of the characteristics of fifteenth-century painting. The inscription reinterpreting the sitter as a sibyl was added at an even later point in time, probably around 1600, to judge by the style of the letters.
(...)
"The assumption that the depiction of Isabella in the Getty Museum was executed around 1450 is confirmed by the results of the dendrochronological examination of the panel, which was fashioned from two oakwood boards. Peter Klein moreover succeeded in proving that one of the two boards came from the same tree as boards used in four further paintings by Rogier and his workshop. The works in question are the two small panels in Vienna showing the Madonna and Saint Catherine, as well as the Lamentation and the Reading Man, both in the National Gallery in London."

No apêndice em que abordamos as aplicações da dendrocronologia, o leitor poderá encontrar mais considerações sobre a datação do retrato de Isabel a partir das outras pinturas acima referidas.