Patient twin: Why Sophia is no longer afraid of cancer

Curious to learn more? Find out how the future of cancer therapy could look thanks to developments in medical technology. Follow the journey of our fictitious patient, Sophia. While our innovation team, of course, considered many different types of cancer and treatment options, only ONE possible scenario is presented here to illustrate how a future product and solution portfolio could improve cancer care.

Sophia has been having a tough time. Two years ago, the 58-year-old was diagnosed with lung cancer – more specifically, with non-small-cell lung cancer (NSCLC). [1] A few months ago, further metastases, so-called oligometastases, formed in her liver and spine – a shock for Sophia and her family.

Initial treatment of all tumors has proven successful. That's why Sophia is now able to not focus on her illness in everyday life. She feels quite reassured. Her health is monitored through regular check-ups with her medical team of oncologists and radiologists, via special apps on her smartphone, and by modern medical technology – all with the help of artificial intelligence, of course.

All medical data collected over the course of Sophia's treatments flow into the digital model of the patient: her digital patient twin. AI continuously evaluates her data in real time so that it can make predictions about the development of Sophia's health. Large language models help to structure disordered data and convert this information into medical reports in language appropriate for Sophia's family doctor, her oncologist, or for Sophia herself.

The data are also used in preparing the medical equipment that Sophia needs for her examinations, applying the optimal settings for her individual needs. This not only results in more precise treatment, but also saves time and capacity for medical staff.

The digital patient model knows the course of Sophia's treatment and the onward plan that she has agreed with her doctors. It knows Sophia's personal preferences and family situation. Via an app on her smartphone, the digital model reminds Sophia when an examination or treatment is coming up. Based on scientific recommendations, her digital twin offers tips and tricks on what she can do before, during, and after a particular treatment to enhance her general condition and well-being.

The data from Sophia's digital patient twin not only help Sophia herself; she has agreed to donate her data ─ in anonymized form, of course. The data is fed into a giant database, that can be used to train AI algorithms, allowing the AI-based decision-making to be constantly improved.

Artificial intelligence helps medical centers simulate therapeutic successes so that the best next step can be chosen. Early warnings based on the data collected allow medical centers not only to schedule appointments for check-ups, diagnostics and follow-up treatment, but also to predict bottlenecks in therapy capacity and adjust their human resources accordingly.

Once a week, Sophie performs a liquid biopsy in the morning after brushing her teeth, a test aimed at detecting cancer cells in the blood. She draws the blood sample herself by simply pricking her finger.

Today, an alarm has sounded in Sophia's portable biosensor device: the biosensors have detected abnormalities in her blood counts that could indicate that her cancer has progressed. At the same time, the AI sends a warning message to Sophia's medical team and suggests a strategy for further action. Sophia is worried.

The lead oncologist calls Sophia directly and reassures her. Following the AI's recommendation, she decides together with Sophia that the next diagnostic step should be a computed tomography (CT) scan. The AI in the patient twinning app on Sophia's smartphone identifies the best CT device for her examination. The AI checks available resources and arranges an appointment with Sophia for the CT scan, including the exact location and travel options. Sophia is even told who will meet her to carry out the CT scan.

Two days later, Sophia makes her way to her CT scan. Based on the data provided by the patient twin, which has already stored the updated data from Sophia's blood test, the scanner unit is optimally adjusted to her specific needs – even before she enters the medical facility.

The radiology technologist onsite simply has to check and confirm the automatically adjusted settings on the equipment. Radiology technologists could even do this from their computer at home, or while on the go using a tablet. State-of-the-art technology makes it possible to carry out complex medical examinations at any time, even in remote areas or regions where medical staff are in short supply.

The CT scanner that the AI recommended for Sophia performs photon-counting computed tomography. Its advantages include a reduced radiation dose and improved image quality. [4] The scanner's particularly high-resolution and high-contrast images help medical staff to recognize small details and differences in tissues, and thus delineate the tumor invasion. In combination with AI, radiomics allows further key information about the tumor to be obtained from the data.

The AI uses the avatar to present possible therapeutic approaches so that the medical team can quickly decide on further treatment: among other things, the AI has created a digital twin of the new tumor based on genetic patterns in complex calculations involving radiomics.

This procedure reduces the number of tissue samples required to be taken from the patient. Therapy can be provided in a faster, more targeted manner. Based on the digital twin of the tumor and patient, the AI predicts how likely it is that the tumor will respond to the various forms of therapy and with what success. Can cryoablation [6] or radiation therapy currently better help Sophia?

Before AI-optimized workflows were introduced, radiation therapy planning was a lengthy process that often left patients waiting several weeks. With the support of AI, treatment planning and Sophia's radiation therapy can now be done within a few hours on the same day, so no precious time is lost.

The AI plans capacity, coordinates appointments, assists with the detailed planning of radiation therapy, and adjusts the settings of all devices precisely to Sophia's needs based on her patient data.

The AI has calculated that Sophia should receive a total radiation dose of 25 fractions, spread over several weeks. Her first radiation treatment will start immediately. Based on the treatment protocol data, the linear accelerator (linac) settings have already been adjusted to deliver exactly the calculated dose. 

The linac has a built-in imaging function that generates more medical images during treatment. It recognizes how the patient's soft tissues and organs move so that it can compensate for these movements. The sensor technology and AI integrated in the linac take into account Sophia's breathing and breathing pauses during irradiation. With the help of this imaging and AI, the radiation dose can be precisely concentrated on the tumor while sparing the surrounding tissue. During quality control, physicians can see whether the treatment was successful, even while it is still ongoing.

For the second radiation dose a few days later, the AI has already evaluated the digital twin for radiation therapy for the results of the first fraction. It recommends an adjustment of the treatment's prescribed dose. The radiation oncologist then decides whether to accept the proposed adjustment to the radiation dose for the second fraction. After consulting with Sophia, he approves the protocol.

Sophia's liver tumor needed treatment about a year ago. At that time, her doctors selected a transcatheter arterial chemoembolization (TACE) on the advice of the AI.

In conjunction with other key technologies, the AI has also supported TACE, firstly by recommending to the medical team the appropriate guidewires and catheters to use as well as the best embolization material for targeted vascular occlusion. The AI simultaneously checked whether that material was available in the hospital's inventory. For the angiography to embolize the blood supply to the tumor, the AI recommended the correct injection site, access routes, and the best angle for the C-Arm.

Even further in the future, patients like Sophia could perhaps be helped by a surgical angio robot in a specialized medical center. Under constant control and monitoring by the medical professionals, who can intervene remotely at any time, the surgical angio robot could automatically guide the catheter through the blood vessels and inject the necessary drugs as well as materials for the vascular occlusion of the tumor. This would allow difficult, rarely performed and time-critical procedures to be offered in remote, rural areas as well.

If Sophia needs a minimally invasive surgical procedure, the physicians could use virtual reality to stage and simulate the procedure in advance. In this way, they could save valuable time in the operating theater and minimize the risk of complications during the procedure. In the operating theater itself, augmented reality headsets would be used so that doctors have all the important information about Sophia's health literally “in view” at all times.

The information in their headsets would navigate the surgeons through the procedure. Experts who cannot be onsite could be linked in via the headset to give their assessment. During the procedure, AI would provide real-time support with decision-making. In this way, the process and progress of the operation could be simulated and recommendations or warnings could be given at an early stage if the AI detects deviations from the planned procedure or an impending complication.

Sophia feels well cared for: If even the slightest sign of a deterioration in her condition should appear, her digital twin would immediately warn her and the doctors treating her. The AI, together with her medical team, would then find an evidence-based, personalized treatment option for Sophia – and continue to help her manage her health.


We need more than just innovative technological developments and visions if digital patient twins and intelligently networked key technologies are to one day become reality: It will take reliable digital infrastructure, for example, that enables a continuous exchange of medical data. The foundations for this would need to be established in today's healthcare systems and hospitals so that patients and medical staff can benefit from them in the future.