Patenting medical devices in Europe is challenging. First, there is the problem that medical use claims are not permitted for medical devices, whilst simultaneously methods of treatment or surgery using a device are excluded from patentability. However, a recent decision from the Board of Appeal in T 0941/24 demonstrates that it is sometimes possible to navigate the fine line of patentability exclusions to claim the use of a surgical device.
 |
| Medical devices |
Legal background: Exceptions to patentability, clarity, and medical devices
In Europe, methods of treatment by surgery are excluded from patentability (Article 53(c) EPC). Particularly, in G1/07, the EBA found that claims which comprise a step "encompassing" an embodiment which is a method for treatment by surgery are not permittable. However, similarly, second medical use claims are not permitted for medical devices (Article 54(4) EPC). This type of claim is permitted only for "substances or compositions" (IPKat). These prohibitions combined create a problem for MedTech companies. In Europe it is simultaneously not possible to patent the method of surgery using a medical device (i.e. in a method of surgery claim), or a medical device for use in a method of surgery (i.e. EPC 2000 second medical use claim). MedTech innovators are often left with the only option of protecting the device itself, and not its surgical use.
Case background
The case in T 0941/24 concerned European patent application EP 3332729, filed by BrainLab. BrainLab is a Munich-based medical technology company that specialises in image-guided surgery, radiotherapy and digital operating room integration. The patent application related to a medical tracking system involving two sensor devices.
The application described how the invention sought to overcome the limitations of traditional medical tracking systems, which typically rely on bulky and expensive 3D cameras to detect markers attached to tracked objects. In these conventional setups, there is a constant risk that the surgeon or other personnel might obstruct the camera's line of sight. To solve this problem, the patent proposed a system using at least two independently manoeuvrable sensor devices, each equipped with both a marker and a marker detector, such as a camera. Rather than relying on a central camera, the sensors detect each other. A control unit then combines the respective sensor data from both devices. Even if the data from a single sensor is insufficient on its own, combining the data allows the system to accurately determine the relative position of the sensors. This approach, the application explained, reduces the likelihood of obstruction and eliminates the need for an external 3D camera.
During examination, the Examining Division (ED) found that the method claim for the device encompassed an embodiment in which a sensor device was attached to a bone, and that this attachment step was implicit. The claim at issue specified:
"A method of determining a relative position between two sensor devices of a medical tracking system, wherein the sensor devices are independently manoeuvrable and can be positioned in a fixed position relative to targets, comprising the steps of determining, with the two sensor devices, respective sensor data comprising the position of a marker device of another sensor device and transferring the sensor data to a control unit, characterized by determining the relative position between the two sensor devices by the control unit by combining the sensor data."
Citing G1/07, the ED found that the claim, by virtue of being directed to a medical tracking system, encompassed a surgical step and was thus excluded from patentability.
Exceptions to patentability (Article 53(c) EPC)
In its decision, the Board of Appeal took the opportunity to examine the use of the word "encompass" in the G1/07 decision. Interestingly, another recent EBA decision to use the word "encompass" is of course G2/21 (use of post-published evidence to support inventive step). In G2/21, the EBA found that the technical effect relied on for inventive step must be understood by the skilled person to be "encompassed and embodied" by the original disclosure of the application as filed. Following G2/21, there has been much debate as to what is meant by a technical effect to be "encompassed" (IPKat).
The Board of Appeal in the present case noted that if "encompass" meant including any conceivable specific implementation falling under a generic feature, then "virtually all method claims would encompass a surgical step and not be allowable under Article 53(c) EPC" (r. 3.1.4). As such, the Board of Appeal reasoned, additional steps that are, for example, defined only in the dependent claims, are not necessarily encompassed by the independent claim (r. 3.1.7). However, the Board of Appeal noted, the principles of G1/07 could be extended to steps which are claimed implicitly as essential, rather than explicitly (r. 3.1.8).
Applying this to the facts of the case, the Board of Appeal was of the view that the claimed method did not comprise any step of attaching a sensor device to a target, either explicitly or implicitly. The Board of Appeal reasoned that the method could be carried out regardless of whether the sensor was attached to a target, such as a bone, or merely carried by a user. Instead of being attached to a bone, as in the examples, the sensors could instead be attached to imaging devices. Therefore, the Board of Appeal concluded, the step of attaching the sensor device to a bone was not an implicit feature and was thus not encompassed by the claimed method.
The Board of Appeal also considered whether the method comprised moving an exposed bone, another surgical step described in the application's workflows. However, in the Board of Appeal's view, the method concerned "a passive, one-time determination which does not rely on any motion or acceleration by the sensor devices or the targets" (r. 4.3.2). The Board of Appeal thus concluded that the claimed method did not encompass any surgical steps and was not a method of surgery.
The decision under appeal was therefore set aside, and the case remitted to the ED with the order to grant the patent.
Consulting the description to interpret the claims
Notably, to arrive at its conclusions, the Board of Appeal specifically referred to G1/24 (r. 4.1.1). The Board of Appeal particularly noted paragraphs in the description that defined the scope of the invention. Specifically, the last paragraph of the description stated: "It is to be noted that the methods and workflows described herein do not relate to or comprise any surgical step. In particular, attaching a cutting block to a bone and performing a cut are not part of the present invention. This invention solely relates to the step of navigating, tracking and verifying by acquiring and analysing data." This paragraph was in the application as originally filed. This Kat wonders whether it would have been considered similarly influential if it had only been introduced as a description amendment later on. The principles of G1/24 imply that it would have been.
Final thoughts
In this Kat's view, the decision in this case might have been a close one without the language included in the description. Key to the Board of Appeal's decision was the interpretation of the specific terms and steps used in the claims, and whether surgery was necessarily implicated. Crucially, the Board of Appeal also found, contrary to the decision of the ED, that attachment to the bone was not an essential feature of the method. Drafting medical device claims is particularly tricky following G1/24, given that the disclosure in the description may be used to import essential features into the claims. However, as this case shows, the description may also be very helpful for shifting the scope of the claims away from non-patentable subject matter.
Further reading
Reviewed by Dr Rose Hughes
on
Tuesday, May 05, 2026
Rating:
Follow the IPKat on LinkedIn 
It is somwhat surprising to me that the Board did not consider whether the claimed method is a diagnostic method practised on the human body, contrary to A. 53(c) EPC.
ReplyDeleteConcerning your comment "Instead of being attached to a bone, as in the examples, the sensors could instead be attached to imaging devices. Therefore, the Board of Appeal concluded, the step of attaching the sensor device to a bone was not an implicit feature and was thus not encompassed by the claimed method.": In my opinion this interpretation is misleading. The point is that the attaching step is not claimed at all and is also not essential for carrying out the invention. The invention also works without attaching the sensor to an object. Therefore it cannot be argued that the attaching step is essential and hence implicit. If an attachment to an object would be essential and hence implicit, it would also encompass the attachment to a bone and would therefore be a surgical method. This is why claim 8 had been deleted before the oral proceedings taking place and which claimed the attachment step. Apparently the BoA interpreted the deleted claim 8 as comprising the attaching step.
ReplyDeleteI would like first to observe that, when faced with the objection that claim 8 ("each sensor device is attached to a target") related to a surgical method in view of the description, the applicant simply deleted claim 8.
ReplyDeleteWhen looking at § [0022], § [0029], § [0033], § [0036], § [0039], § [0041], § [0042] and § [0044] of the description, it is difficult to follow the board when it considers that the method as exemplified and disclosed neither comprises nor encompasses a surgical method. All the examples in the description relate to a surgical method.
This conclusion is in accordance with G 1/24 requiring the description to be “consulted”. Consulting does not mean ignoring large chunks of the description and only concentrating on the last § of the description.
The last paragraph at the end of the description states indeed that the invention “solely relates to the step of navigating, tracking and verifying by acquiring and analysing data”.
In view of the description and order to clarify the situation, the method claim should have comprised a clear disclaimer as mentioned in the last § of the description. Such a disclaimer is perfectly allowable under G 1/03 or G 1/16. Why has this solution not been adopted?
Since the Board considered that the method "does not comprise attaching the sensor devices to a target, either implicitly or explicitly" there was no need for such a disclaimer. It would appear this comment is insinuating that the Board was not applying the EPC in a correct manner.
DeleteIn Peter de Lange's Just Patent Law Blog , his take on T1239/03, he asks whether the EBA in G1/25 will ground its Decision on the EPC or on something else, namely on:
Delete"a more administrative nature such that the Administrative Council is competent to regulate the matter in the Implementing Regulations,"
Perhaps, these days, it is no longer a sine qua non that the Boards of Appeal at the EPO have to apply the EPC "in a correct manner". Or perhaps there is by now an ever more expansive view as to what manner is "correct"?
Max, the more pertinent question is whether there is any provision in the EPC that clearly and unambiguously requires Guidelines-style adaptation of the description.
DeleteThe absence of clear and unambiguous legal basis would appear to be fatal for a purported barrier to grant. In this respect, it is interesting to note the persistent divergence of views between the proponents of description adaptation. For example, if we include Peter’s preferred basis (Rule 48(1)(c) EPC), this would bring to seven the total of different (combinations of) legal provisions / principles cited by different Boards or individuals as providing legal basis for description adaptation. This seemingly irreconcilable divergence of views would appear to prove beyond any reasonable doubt that the clear and unambiguous standard is not met.
@ Kant-06.05.2026-09.55.00 GMT+1
DeleteI am not insinuating anything. It is a matter of fact that the description is full of references to medical methods using the medical device of claim 1.
I do not deny that the method can be applied outside the medical field, and there is sufficient material for this in the original disclosure.
If it is not possible to distinguish in a claimed method, effects not falling under Art 53(c) from effects not falling under the prohibition of Art 53(c), then method is not allowable. This is nothing new at the EPO, see e.g. T 290/86 or T 820/92 and the like.
In Point 2.1 of the order of G 1/03, I read “A disclaimer may be allowable in order to disclaim subject-matter which, under Art 52 to 57, is excluded from patentability for non-technical reasons. This was confirmed in G 1/16.
In the present case, the method has medical and non-medical applications. It is thus correct to refer to the possibility of a disclaimer under G 1/03 or G 1/16, rather than resorting to an interpretation of the claim ignoring G 1/24. .
When consulting the description as required by G 1/24, the necessary conclusion that the bord has had a rather selective reading of the latter. It carefully ignored the passages relevant to medical methods and concentrated on the last § of the description.
When looking at all decisions applying G 1/24, even T 439/22, the use of a disclaimer imposes itself.
Do you want a holistic approach when interpreting a claim under G 1/24, or do you favour a “selective” holistic approach?
@ Max Drei-06.05.2026-12.42.00 GMT+1
DeleteIn Peter de Lange's Just Patent Law Blog, there is no take on T1239/03 to be found. This decision is also not found in Peter de Lange's amicus curiae for G 1/25. Could you please check your source.
Proof of the pudding-06.05.2026-15.40.00 GMT+1
DeleteIt is not because there are different views on the legal basis for requiring adaptation of the description among the proponents of the adaptation of the description, that the EBA will be unable to indicate the legal basis requiring adaptation of the description.
The postulate according to which there is not “any argument all to prove beyond any reasonable doubt that the clear and unambiguous standard is not met”, is therefore to be refuted. You are manifestly underestimating the EBA.
For your information, I invite you to look at T 235/25-2 published today. As we are near the OP in G 1/25, I find this decision rather interesting.
The description in AR6 and 6b was not adapted to the claimed deemed allowable by the board in a first decision T 235/25-1. A second OP was held, only for discussing the adaptation of the description.
In this decision the board found an inconsistency between claims and description and considered that Art 84 was not met. A post on this decision will appear tomorrow on my blog.
In the last part of Reasons 4, the board held that the inconsistency between the claims and the description leads to uncertainty as to which features form part of the invention and which are merely optional.
The patent was only maintained on the basis of AR6c. In AR6c, the optional character of features limiting the independent claim was deleted from the description.
In T 235/25-2 the board found Art 84 as basis for requiring the adaptation of the description. There is thus at least one legal basis possible. This was the legal basis proposed by most, if not all those in favour of adaptation of the description.
@DXT
DeleteMethinks that you doth protest too much.
The relevant issues here are the established facts and logical deductions from those facts.
The established facts are that:
1) the text of EPC1973 was finalised over 5 decades ago;
2) that text includes all of the (precursors) of the provisions that are now touted as candidates for legal basis for mandatory adaptation of the description;
3) there are over 4 decades of case law which interpret the provisions of the EPC; and
4) despite all of points 1 to 3, there is still no general (if not universal) agreement upon which, if any, provision(s) of the EPC provide legal basis for mandatory adaptation of the description.
The logical deduction from points 1 to 4 is that the EPC does not contain clear and unambiguous legal basis for description adaptation: as if it did, then the proponents of description adaptation would long ago have reached a consensus view on where that legal basis can be found. What we have instead is a persistent divergence of views which rely upon mutually incompatible theories regarding purported legal basis.
Whilst I have no doubt that members of the EBA are very knowledgeable and experienced, they are not so omnipotent that they can magically identify clear and unambiguous legal basis that has eluded us all for over half a century - not least because this would require the EBA to turn established facts on their head and to defy logic.
Regarding T235/25, I am surprised that you see this as pertinent in view of the President's rather dubious decision to continue proceedings where the decision depends entirely upon the outcome of G 1/25. I would have thought that you would instead be more concerned about whether the Board of Appeal is overriding its obligation not to issue a decision that prejudges the EBA's ruling.
Will now have to read your blog post, DXT, and T235/25. But my initial reaction is that it provides a clue which sort of inconsistency offends Art 84 EPC, namely that which renders it unclear which features of "the invention" are mandatory and which are merely optional.
DeleteWhen I was in training, back in the 1970's, it was explained to me that it was the job of the claim drafter, me, in service to the client looking for a scope of protection commensurate with ther contribution to the art, to sort features into two categories, those that define the invention in the claim, and those further features that are merely optional. The former go in the independent claim, the latter not.
Fifty years of practice later, I have difficulty with the notion that an inconsistency is by itself enough to render unclear the claim that was so many thoughtful hours in the drafting. Caveat: if the descrn includes an explicit definition of a feature in the claim, that definition should be imported into the claim. Of course. so the claim scope is clear.
@DXThomas
Delete"In Peter de Lange's Just Patent Law Blog, there is no take on T1239/03 to be found."
Yes there is: https://justpatentlaw.blogspot.com/2026/05/t-123903-and-g-125.html
My post of 5 May indeed.
Delete@ Proof of the pudding-07.05.2026-09:43:00 GMT+1
DeleteIt is not the first situation in which I can return your compliment. If I am meant to protest too much, what are you actually doing? Ever heard about the straw and the beam in the eye?
You come to the logical deduction from your points 1 to 4 that the EPC does not contain clear and unambiguous legal basis for description adaptation: as if it did, then the proponents of description adaptation would long ago have reached a consensus view on where that legal basis can be found. You will not be surprised that I disagree.
There is no need for the very knowledgeable and experienced EBA members to have to resort to magics in order to identify a clear and unambiguous legal basis for requiring (or not) adaptation the description. Foremost it is Art 84. R 42 and R 48 could complement the requirements of Art 84.
This topic was discussed extensively this morning during the OP for G 1/25.
When in the case leading to T 56/21, the board had announced intention to file a referral to the EBA, my reaction was to say that the board will not be welcomed by the EPO’s and the BA’s management as it would mean more or less stalling most of the procedures before first instance divisions and the boards.
The decision of the president’s to continue the procedures started in spite of referral G 1/25 did thus not come as a surprise to me.
As far as T 235/25-2 is concerned, it is left to the discretion of the board to stay the proceedings or not. There is no obligation to that effect derivable from the EPC or any ancillary regulations. However, I would say that the board having made a referral to the EBA should indeed stay the proceedings if a similar case has to be discussed before it.
What ccould be the legal basis obliging a Board of Appeal to stay the proceedings and obliging it not to issue a decision that could prejudge an EBA ruling?
By the way, this is exactly what the board did when it gave up the idea of a referral on the adaptation of the description and issued T 56/21. It did not stay the proceedings, although it must have been fully aware that its considerations could be impacted by the issue in G 1/24. But T 56/21 is a decision, you and plenty of others, all against adaptation of the description, use like a life buoy to argue against the adaptation of the description.
Without T 56/21, and the like, it would have been much more difficult to argue against adaptation of the description.
@ Max Drei-07.05.2026-10.23.00 GMT+1
DeleteI am surprised that fifty years of practice later, you have difficulty with the notion that an inconsistency is by itself enough to render unclear the claim that was so many thoughtful hours in the drafting.
As well as assessing non-obviousness does not imply any judgement of value, be it about the creative effort of the inventor, or be it about the importance or the usefulness of the invention, deciding whether the matter for which protection is sought is defined correctly in the claims and in the description does not depend on the efforts of the drafter and the hours he spent in drafting.
The description should not contain any statement which allows doubts about the matter for which protection is sought. It is as simple as that.
Oh the joy of it all, in the context of G1/25, adaptation of the description to the subject matter defined by the claim. Next thing I do, take a look at the Decision to see whether the order to grant a patent includes a command to the ED first to ensure that the description is adapted to the allowable claim. I mean, must the Examples now be explicitly tagged as "not" the invention? Or is the order to grant with the description unchanged from the one considered by the Board.
ReplyDeleteAnd another thought: could Applicant have avoided the refusal, and the journey to the Board of Appeal and back, and all the expense and delay, simply by conforming the description earlier in the prosecution?
Or have I overlooked something obvious? If so, do tell.
Moee joy. On reading the description proposed by the ED for grant, I find multiple instances of amendment from "embodiment of the invention" into "embodiment of the disclosure". I presume that this amended form of words accommodates not only a set of claims lacking independent method claim 7 but also, just as well, a set of claims that includes the controversial method claim. Ha!
ReplyDelete@ Max Drei-06.05.2026-12.13.00 GMT+1 and 12.28.00 GMT+1
DeleteIn its order the board decided that the patent should be granted with “Pages 1 to 23 as indicated in the examining division's intention to grant dated 22 December 2023”
The board seems thus to have endorsed the amendments proposed by the ED.
One of the interesting points to be dealt with in G 1/25 is indeed what should happen with embodiments contained in the original disclosure, but which do not any longer correspond to the definition of the matter for which protection is sought once the claim has been limited/amended.
In a post of 27.04.2026, I summarised the 5 situations in which, in my humble opinion, an adaptation of the description appears to be necessary.
The optional features are dealt with under case 3 and the embodiments not any longer covered by the claims are dealt with under case 4.
In response to DXT's request, here is a Link to the Just patent Law blog and its mention of Decision T 1239/03
ReplyDeletehttps://justpatentlaw.blogspot.com/2026/05/t-123903-and-g-125.html
@ Max Drei-06.05.2026-20.55.00 GMT+1
DeletePlease accept my apologies for not finding Peter de Lange’s post on T 1239/03. I used an outdated link.
T 1293/03 is the decision used by the Anonymous in its TPO of 03.04.2026 to accuse the Rapporteur D. Rogers of partiality. This grave and defamatory accusation is completely without foundation. The rapporteur simply reported about the decision without any judgement of value. As you commented the corresponding post on my blog, you knew this very well.
It should be reminded that T 1239/03 it is the proprietor which, during the OP before the OD, voluntarily deleted two examples. The board noted that the proprietor by deleting Examples 2 and 3 intended to have the percentage of ethylene content interpreted as weight percent, i.e. to reach a different interpretation of the claimed subject-matter. No wonder that the proprietor was then faced with an objection under Art 123(2).
T 1239/03 has been issued well before G 2/10 and G 1/24, drawing from T 1239/03 conclusions allegedly valid for G 1/25 is, to say the least, incorrect.
A voluntary deletion in order to achieve a different interpretation of the claim is not to be compared with marking embodiments as not any longer corresponding to the definition of the matter for which protection is sought. Claiming that the two situations are comparable does not appear to be correct.
I maintain my view that embodiments not any longer corresponding to the definition of the matter for which protection is sought should not be deleted, but marked accordingly. By doing so, neither added matter nor increasing the scope of protection can be discerned.
As far as the part of Peter’s blog you mentioned, i.e. "a more administrative nature such that the Administrative Council is competent to regulate the matter in the Implementing Regulations", I allow myself to say that it is very selective not to say biased.
It does not appear to me that the AC should necessarily amend the present implementing regulations. With Art 84, R 42 and R 43 as they are, the EBA is well in a position to tke a decision in G 1/25.