Accuracy research of the Three-dimensional visualization imaging system for preoperative imaging in malignant tumors of the head and neck

Imagine trying to navigate a dense, foggy forest to find a hidden treasure. In the real world, surgeons face a similar challenge when operating on head and neck cancers. The "forest" is the complex anatomy of the human neck, packed with vital roads (blood vessels), power lines (nerves), and the "treasure" (the tumor) that needs to be removed without damaging anything else. For decades, doctors have relied on flat, two-dimensional maps—like looking at a single slice of a loaf of bread to understand the whole cake—to plan these surgeries. While helpful, these flat maps can be tricky to read, often leaving surgeons to guess how deep a tumor goes or exactly where a lymph node (a tiny filter in the body that can catch cancer cells) is hiding. This is where a new kind of technology steps in: 3D visualization. Think of it as turning that flat map into a fully interactive, rotatable hologram that you can spin, zoom into, and explore from every angle, giving a much clearer picture of the terrain before the surgery even begins.

This study, conducted by researchers from Xuzhou Medical University and the University of Malaya, puts this "hologram" technology to the test. They wanted to know if their new 3D system, called 3D-VIS, could accurately predict the location and size of cancerous lymph nodes in patients with head and neck tumors, compared to the actual findings during surgery. The team took CT scan data from 34 patients and fed it into their AI-powered system, which built detailed 3D models of the neck. They then compared these digital models against the real-life measurements taken by surgeons in the operating room. The results were impressive: the 3D system was remarkably accurate. It correctly identified where abnormal lymph nodes were located in almost every case, and the size measurements from the 3D models were nearly identical to the physical measurements taken during surgery. In fact, the predicted volumes of the lymph nodes matched the actual volumes with such precision that the difference was statistically negligible. The study suggests that this technology could become a powerful tool for surgeons, helping them plan operations with greater confidence, potentially reducing surgery time and improving outcomes, though the authors note that more testing with larger groups of patients is needed to confirm these benefits across the board.

The Digital Twin of the Neck

Imagine you are a surgeon about to perform a delicate operation on a patient's neck. The neck is a busy intersection of highways (blood vessels), tunnels (airways), and tiny checkpoints (lymph nodes). If a tumor is hiding there, it's like a criminal hiding in a crowded city. Traditionally, doctors have used 2D CT scans to find the criminal. But looking at a 2D scan is like trying to understand a 3D city by looking at a single, flat street map. You can see the streets, but you might miss how tall the buildings are or exactly where a side alley leads.

To solve this, the researchers developed a "3D Visualization Imaging System" (3D-VIS). Think of this system as a magical scanner that takes the flat CT images and builds a fully interactive, three-dimensional "digital twin" of the patient's neck. This isn't just a pretty picture; it's a data-rich model that can be spun, zoomed, and sliced open on a computer screen. The system uses artificial intelligence (AI) to automatically find and highlight the tumor and the lymph nodes, separating them from the surrounding muscles and blood vessels.

The Great Test: Digital vs. Real

The researchers didn't just build the model; they wanted to know if it was trustworthy. They gathered 34 patients with head and neck cancers who were scheduled for neck dissection surgery (removing lymph nodes to check for cancer spread). Before the surgery, the team used the 3D-VIS to create a model of each patient's neck. They measured the lymph nodes in the model, noting their size, shape, and exact location.

Then, the surgeons went into the operating room. They removed the lymph nodes and measured them again, this time with a ruler (a vernier caliper) and by using the water displacement method (a classic science trick to measure volume) to get the "ground truth" of how big the nodes really were.

The big question was: Did the digital model match the real thing?

The Results: A Near-Perfect Match

The answer was a resounding "yes." The study found that the 3D system was incredibly accurate in two main ways:

1. Location: The system correctly identified where the abnormal lymph nodes were in the neck. When they compared the 3D map to the actual surgery, the agreement was extremely high. For example, in the different zones of the neck (called Levels I through V), the 3D system predicted the presence of nodes with a consistency score (kappa) of over 0.8, which is considered "excellent" agreement. In some zones, the match was perfect (1.000).
2. Size and Volume: This is where the magic really happened. The researchers compared the size of the lymph nodes in the 3D model to the size measured during surgery.
   • Diameter: There was no significant difference between the 3D measurements, the standard CT scan measurements, and the actual surgical measurements. They were all basically the same.
   • Volume: The 3D system predicted the volume of the lymph nodes with stunning precision. The average predicted volume was 1.32 cm³, while the actual volume measured in the operating room was 1.34 cm³. The difference was tiny—only about 0.09 cm³ on average.
   • The Math: When they ran the numbers, the relationship between the predicted volume and the actual volume was almost a perfect line (a correlation coefficient of 0.997). This means if the 3D system said a node was a certain size, it was almost guaranteed to be that size in reality.

Seeing the Invisible

One of the coolest parts of the study was how the 3D system handled blood vessels. In some patients, the tumors were squishing or twisting the major veins in the neck. The 3D models showed these deformities clearly, letting the surgeons know exactly where the "roads" were blocked or bent before they even made an incision. In one case, the system showed a vein running right through a tumor, a detail that might have been missed on a flat scan.

The researchers also tested how well the system could help patients understand their own condition. They generated a QR code that patients and their families could scan with their phones. This allowed them to see the 3D model of the tumor and lymph nodes on a smartphone, spinning it around to see exactly what the surgeons were talking about. This made the pre-surgery conversations much clearer and less scary for the families.

What This Means (and What It Doesn't)

The study concludes that the 3D-VIS system is a highly accurate tool for planning head and neck cancer surgery. It suggests that using these 3D models can help surgeons plan better, know exactly where to cut, and avoid damaging important structures.

However, the authors are careful not to call this a "solved problem" for every single aspect of surgery. They noted a few limitations:

• Primary Tumor Size: While the system was great at measuring lymph nodes, they didn't statistically prove it was perfect at measuring the main tumor's exact volume. They explained that this is partly because oral tumors can be hard to see clearly on CT scans due to metal fillings in teeth, and it's hard for surgeons to measure the exact edge of a tumor while cutting it out.
• Sample Size: The study only looked at 34 patients. While the results are promising, the authors say that larger studies with more patients are needed to be absolutely sure this works for everyone.
• Single Center: All the surgeries were done by one team at one hospital. Future studies would need to see if other surgeons in other places get the same great results.

In short, this paper shows that turning flat medical scans into interactive 3D holograms is not just a cool trick—it's a highly accurate way to map out the battlefield before the fight begins. It suggests that in the near future, surgeons might rely on these digital twins to navigate the complex forest of the human neck with confidence, ensuring they remove the bad cells while leaving the good ones untouched.