Unofficial comments on FDA's RASD discussion paper

A few weeks ago FDA released a Discussion Paper to facilitate the dialog of the upcoming Public Workshop (starting on Monday).

The document is a great starting point, and identifies key critical aspects of the field, however, it has some shortcomings that should be addressed to make the meeting and the overall dialogue most fruitful. 

1.  The FDA Discussion Paper defines robotic surgery (p.2) as a classical master-slave teleoperational system. Later it confirms that it is only addressing "da Vinci type" robotic devices as RASD (p.8). This narrowing of the scope immediately creates a vast void on the imagined landscape of robotic surgery devices, since there are already a dozen different  systems cleared by the FDA (see our recent survey), and numerous other types are to come. Image-guided surgical robots (ROSA, neuromate, THINK Surgical (aka ROBODOC), iSYS, Mazor's SpineAssist), the catheter robots and other special types (e.g., ARTAS' hair restoration system) are all considered to be surgical robots. These systems' milestone achievements are all missing e.g. from the historical overview (p.4).
2. Further NOTES devices, capsule robots, nanorobots and other emerging platforms should also be addressed by the workshop. 
3. Energy delivering robots are not addressed either, such as CyberKnife or HIFU robots.
4. While FDA uses the current ISO definition of Robots (ISO 8373:2013), this standard also defines "robotic devices", through which the scope of the definition can be extended. Most importantly, the ISO/TC 184/SC 2/JWG 9 is actively working on a Technical Report to define the "Degree of Autonomy", which is currently not defined. According to the current draft, kinematic capabilities (and master-slave teleoperation in general) will be considered to be a lower level of DoA, therefore da Vinci type systems are expected to clearly fall under the category of "robot".
5. Haptic feedback (p.5) exists for some systems (e.g., MAKOplasty), yet the cost/benefit ratio should be established first for master-slave systems.
6. Within design (p.6), usability is a key issue. The human-machine interface has a major importance in the case of master-slave systems, and current relevant standards are not detailed enough. 
7. Standardization of robotic surgery training has advanced significantly in the past few years, this should be better presented in the paper: the R-FLS is now becoming a validated curriculum. 
8. App1: Challenges and Opportunities1: since there yet exist no strict standard for any particular surgical procedure (only schools and basic protocols), it is extremely hard to test any RASD system against anything else than its own design specification.
9. Testing a device is not enough, RASD should also be considered as a cyber-physical system, human and machine together! 
10. Ch&Opp2: Medical imaging has gained a high level of standardization. Surgical planning and then executiong should also follow a similar path. 
11. Ch&opp6: The technology transfer from academia to industry should be streamlined to enable better systems coming to the market. In the mean time, the fundamental problem of "competition vs. collaboration" affects largely the field, since the existing companies are profit proven. The value for patients vs. the value for shareholders must be balanced. 

Minor comments:

1. Minor notice that the 3 components mentioned as key parts of a RASD (p.2) are only visible as 2 in the newer generation. 
2. The K965001:1997 cited (p.4) refers to the "Mona" system.  

The statements above do not represent any official standing points or arguments, and solely form the professional opinion of the author of this post.