Developing AI in Combat Healthcare

 By Ajinkya Jadhav


Indian Army has been in active combat for the past three decades primarily owing to the asymmetric threat it has been countering. It has fought four wars in the past and is battle-hardened. The emerging conventional threat along both the Northern borders and the unconventional threat owing to the situation in Afghanistan demands a high state of combat readiness. The army operates in extremely inhospitable terrain and in adverse weather conditions. The terrorists employ guerrilla tactics and take advantage of both the terrain and weather to combat the security forces. The combat operations especially in counter-terror operations have taken a toll on the security forces. The two main battle winning factors in any combat are superior training and high quality of weapons and equipment. However, in combat when there is an active firefight from both sides, casualties are imminent. Rescuing wounded soldiers during active battlefield situations is one of the most dangerous positions and the cause of many military deaths. According to researchers, around 86% of battlefield deaths occur during the first thirty minutes post-injury[1]. This is the reason researchers across the world are working toward developing artificial intelligence (AI) technology that can bolster battlefield healthcare. This article analyses the need for developing artificial intelligence technology in battlefield healthcare in Indian conditions in order to minimise the casualties of our security forces and ensure they live to fight another day, thereby boosting the morale of the Indian Army.

First Aid at Unit Level

Providing advanced healthcare to the soldiers on the battlefield at the site of the combat operation especially in the first 30 mins is extremely vital. The difficult environmental conditions and geographical separation between the injured warrior and healthcare providers add to the challenge. The immediate first aid currently relies on the intuition and experience of the battlefield nursing assistant (BFNA)/Regimental Medical Officer (RMO)/medically trained combat soldier, his medical skill, availability of life-saving equipment and medicines. But the most critical step is the correct assessment and the initial diagnosis based on which a decision has to be taken to administer the first aid. The US Army based on its experiences in the global war on terror is developing an AI-based interoperable Algorithms for Care and Treatment (iACT) program that would feature artificial intelligence algorithms that generate indications, warnings and suggestions for soldiers and medics, giving them improved abilities to monitor, diagnose, triage and treat fellow service members[2]. Indian Army has rich operational experience in combat in conventional and counter-terror operations, there is a considerable amount of data available in administering lifesaving treatment immediately on the occurrence of casualty.  AI modules that integrate algorithms and large-scale medical databases in a machine learning-based clinical decision support system can be specifically developed to provide critical medical support by our frontline troops and medics. Accessed datasets should include large numbers of the past medical trauma cases, including patient diagnoses, complete vital signs sets, medications given, medical treatments performed and patient outcomes. The AI-based first aid module at the unit level should analyse manually entered data as well as that received from vital signs monitors and, drawing on its decision-support functions, to provide recommendations for medical treatment. This AI module should have predictive capabilities, giving soldiers/BFNA, AI-based alerts forecasting injury patterns and when patients might deteriorate. AI modules can be built into small data tablets enabled by a network of Software-defined radios that can help in clinical diagnosis and recommend suitable first aid to the injured soldiers in the operational area.

Casualty Evacuation

The next step is the evacuation of the casualty either to a field hospital or to the nearest military hospital by air/road. The most challenging part is negotiating rugged terrain and transporting the casualty on stretchers. The golden hour principle ie shifting the casualty to the nearest hospital within an hour could be a life and death situation for the injured soldier. US Army is developing battlefield casualty-extraction robots, designed especially for retrieval of injured soldiers which can negotiate uneven terrain while carrying the weight of a human. Metal-bodied remote-control UGVs currently have a top speed of 10 km per hour (6.2 miles per hour) and can easily lift weights up to 227 kg (500 pounds)[3]. The Army casualty evacuation process can be automated with the help of AI tools. The process can include automation of location, dispatching decision-making requires accurate data, valid analytical techniques, and the deliberate integration and ethical use of both. Artificial intelligence and more specifically, machine-learning techniques combined with traditional analytic methods from the field of operations research provide valuable tools to automate and optimise the casualty-evacuation location and dispatching procedures.  A machine learning module can be developed based on the data available in each sector which can be built by existing data on the nature of casualties, priorities, helipads, weather conditions, terrain details, medical facilities available, first-aid data sets, time and distance etc. A predictive AI module can enable locating and prioritising casualty evacuation resources optimally.

Advanced Healthcare

Advanced healthcare is provided at the military hospitals where a casualty is transported after provisioning of first aid either by air or road. A highly skilled team of doctors treat the patients. AI-based modules can be developed to assist them with inaccurate diagnosis, correct decision making and proper line of treatment. AI tools can be helpful in the conduct of complex surgeries and providing advanced lines of treatment to the doctors treating casualties in hospitals[4]. The introduction of Cyber Medicine enables physicians and soldiers to utilise technologies including mobile apps, robotics, connectivity to wired/wireless networks, satellites, clouds, HPCs, and software to identify, assess and treat the injured[5]. The US Army has collected a vast amount of data on soldier injuries, data that currently goes to the DOD trauma registry, and using it in a way that paints a predictive picture of future battlefield events via machine learning. It has built a Trauma Directory that has data of very specific combat injuries in very specific locations against specific enemies the AI algorithms have been potentially used in machine learning applications, the data would have to be cleaned and structured to be useful in training machines[6]. The second model is developed by the US Army is getting better data from soldiers, both before they’re on the battlefield and during, by shrinking the size of soldier-worn sensors and enabling those sensors to collect more data[7]. The Indian Armed forces have one of the finest and experienced doctors, they have displayed sterling capabilities in all operations to save lives. Their experience needs to be utilised in machine learning and assist future generations in correct and accurate decision making.


  • Indian Army needs to develop an AI-based prototype for critical sectors especially the ones conducting counter-terror operations at the battalion level for monitoring, diagnostics, and intervention with an aim to provide life-saving techniques and strategies for trauma care on the battlefield.
  • A casualty evacuation, AI module could be developed to prioritise and optimise the casualty evacuation resources facilitating swift decision making by the commanders and staff.
  • The military hospitals in the field locations need to develop multiple AI modules to enhance the capabilities of the team of doctors to correctly diagnose and take timely decisions to administer a proper line of treatment coupled with building robotic capabilities to assist in life-saving surgeries.
  • These AI modules are softwares and can be developed indigenously in a sectorised model based on the threat. They can be executed at the formation level initially then scaled up depending on the progress and development. AI engines mature over a period of time depending on the data sets utilised for machine learning. There is no dearth of experience in the Indian army that data can be utilised to build these modules.


The security situation in the Indian subcontinent is critical, belligerent adversary along the Northern borders is constantly challenging India. The volatile situation in the Af-Pak region is likely to have an influx of terror on Indian soil. The future combat is likely to be more lethal and the Indian Army needs to be prepared for these challenges. Harnessing new generation disruptive technologies to enhance our capabilities in combat is vital to our success in operations. AI is changing the face of the modern battlefield: AI is and will continue to be used not only for offensive measures but also is playing a significant role in defensive measures, including saving lives on the front line. Its time the Indian Army addresses this critical issue and takes logical steps to adopt this technology.


[1]  Nichole Heydenburg. (2019), ‘How AI can help in battlefield healthcare’, Military embedded systems, 11 June 2019, available at

[2]  GCN(2021), ‘Data, AI to power medical support on the battlefield’, GCN publications, 24 March 2021.  available at

[3]  Heydenburg Nichole (2019), ‘How AI can help in battlefield healthcare’, Military embedded systems.

[4] Micheal Matheny (2019), ‘Artificial Intelligence in Health Care:The Hope, the Hype,the Promise, the Peril’ National Academy of medicine, Washington paper, 2019, available at

[5] J. S. Farroha, B. S. Farroha(2019), ‘Enabling intelligent battlefield healthcare through secure cyber medicine’ Proc. SPIE 11015, Open Architecture/Open Business Model Net-Centric Systems and Defense Transformation 2019

[6]  Ibid

[7] Ibid.