Page 171 - Read Online

Page 64 Endo et al. Art Int Surg 2024;4:59-67 https://dx.doi.org/10.20517/ais.2024.09

five groups based on age, tumor size, extent of hepatectomy, and CA19-9 levels. This personalized approach
to determining margin width may assist in identifying patients who would benefit most from a wider
negative resection margin. Moreover, using such a tool may help avoid unnecessarily wider margins,
preventing radical surgery and subsequent surgical morbidity.

Operative aids for HB surgery
AI holds substantial promise in HB surgery, especially related to potential intraoperative support. Previous
studies have highlighted the potential usefulness of AI in laparoscopic cholecystectomy [61,62] . For instance,
Madani et al. demonstrated the efficacy of AI technology employing deep learning algorithms to identify
safe and hazardous zones of dissection and other anatomical structures during laparoscopic
[62]
cholecystectomy, improving the performance of operating surgeons . Intraoperative AI can assist
surgeons, particularly trainees, in decision making during surgery and help maintain quality control, as well
as facilitate training efficiency.

Accurate assessment of the future liver remnant (FLR) volume is widely acknowledged to reduce the risk of
post-hepatectomy liver failure . To mitigate this complication, it is crucial to calculate the precise volume
[63]
of FLR preoperatively and plan a surgical approach accordingly. Therefore, there is a need to improve
conventional methods for this calculation (i.e., contrast-enhanced CT scan). Winkel et al. developed a
CNN-based algorithm that demonstrated good accuracy, speed, and agreement with manual
[64]
segmentation . This approach could potentially improve the quality of 3D reconstruction of the liver,
which may help more accurately estimate the FLR [65,66] . Incorporating techniques such as augmented reality
(AR) and mixed reality allows for the synchronization of 3D-reconstructed images with real-time surgery,
representing a safer and more reliable surgical navigation method. Notably, Ntourakis et al. reported in a
pilot study that AR aided in detecting missing lesions post-chemotherapy for colorectal liver metastases,
resulting in a higher likelihood of a margin-negative resection without any local recurrence . The
[67]
application of AR in robotic hepatectomy also has the potential to enhance a surgeon’s ability to achieve a
safe tumor resection with an adequate margin .
[68]

Future perspectives and potential challenges
Looking ahead, the future of AI in healthcare holds promise, yet significant challenges remain. Addressing
knowledge gaps surrounding data quality, data governance, interoperability, and algorithm transparency
will be paramount. Researchers will need to focus on developing robust frameworks for data integration,
[69]
standardization, and ethical AI deployment . Additionally, efforts to enhance the interpretability and
accountability of AI algorithms will be essential to foster trust among healthcare professionals and patients.
The success of AI integration into the clinical setting will be related to external validation in multiple
cohorts. In addition, the applicability of AI can be hindered by the dearth of an easy-to-use interface for AI-
based models. Over the next five years, we anticipate continued progress in AI-driven diagnostics,
personalized medicine, and surgical innovations. However, realizing the full potential of AI in healthcare
will require collaborative efforts across academia, industry, and regulatory bodies to ensure responsible and
equitable implementation while maximizing patient outcomes.

CONCLUSIONS
Recent advancements in AI offer the chance to enhance the care of patients, as demonstrated in the current
study that highlighted the integration of AI into the care of patients with HB tumors. Specifically, AI models
have the potential to impact patient stratification and decision making and are poised to become integral
components of future surgical research and care. As these technologies continue to evolve, their application
could revolutionize medical practices, introducing an era of more precise diagnostics, personalized